Around 16.9 million people suffer a stroke each year and the global burden of stroke is increasing dramatically [1]. China has the biggest stroke burden in the world and about 70% of cases are due to ischemic stroke, when a blood vessel supplying the brain becomes obstructed [2]. Therefore, ischemic stroke is a major cause of morbidity and mortality for Chinese adults. Atrial fibrillation (AF) is an important risk factor for an acute ischemic stroke [3]. Strokes related to AF are also associated with higher mortality and more severe neurologic disability [4]. So, patients with AF are encouraged to consider prevention measures such as oral anticoagulation therapy and modifying their diet and behavior [5].

Dyslipidemia is another well-established risk factor for stroke [6]. Dyslipidemia is implicated in the development of atherosclerosis and coronary heart disease, which are both closely related to AF development [7]. However, while high levels of low-density lipoprotein cholesterol (LDL-C) and low levels of high-density lipoprotein cholesterol (HDL-C) are closely associated with developing coronary artery disease, [7] the relationship with AF is less obvious. Meta-analysis suggests an inverse relationship between serum total cholesterol (TC), LDL-C, and HDL-C levels and AF risk, although there was no significant association between triglyceride (TG) levels and incident AF [8]. However, because both AF and dyslipidemia are independent risk factors for stroke there is a theoretical possibility that patients who experience both may be at greater risk of ischemic stroke.

One study, identified a high level of LDL-C as an independent risk factor for stroke in patients with AF [9]. Another study found that statin use by patients with AF prior to stroke reduced oxidized low-density lipoprotein levels and improved outcomes after stroke [10]. However, large-scale studies are lacking, so it is not clear whether blood lipid levels should be lowered in all patients with AF and if so, the appropriate level of blood lipids needs to be established with quantitative evidence.

Ischemic stroke is a major adverse event in patients with AF, which affects the prognosis of patients [11]. If AF is combined with dyslipidemia, there is a theoretical superposition effect, which further aggravates the occurrence of ischemic stroke [12]. In addition to directly causing ischemic stroke, blood lipid levels also reflect the level of inflammation in patients [13, 14], and studies have shown that blood lipid levels are related to the occurrence of atrial fibrillation [15]. Therefore, in addition to the superposition effect of AF with dyslipidemia, there may also be a synergistic effect, by which this combination further increases the incidence of ischemic stroke. Blood lipid is a risk factor for stroke, but medical personnel primarily focus on controlling the ventricular rate and anticoagulation in patients with AF. Elevated blood lipid levels can cause endothelial inflammation, theoretically increasing the likelihood of forming thrombosis. Thus far, it is unknown whether or not patients with AF need blood lipid management.

Therefore, our study aimed to explore the correlation between baseline blood lipid level and the occurrence of ischemic stroke in patients with AF. The China Atrial Fibrillation Registry (China-AF) is a prospective, multicenter, hospital-based registry of participants diagnosed with AF. This registry, therefore, provides a large amount of real data that can be analyzed and explored, to provide reference levels and evidence for the control of blood lipids in patients with AF.

In order to discover whether blood lipid levels are associated with ischemic stroke in patients with AF, this study investigated baseline blood lipid levels and other clinical factors in patients with AF who experienced ischemic stroke and compared them with the values in matched patients with AF who did not experience a stroke. The results show that TG levels were higher in the stroke group than the non-stroke group, but TC, LDL-C and HDL-C were similar. Elevated TG and HDL-C was also associated with ischemic stroke. Subgroup analysis suggested TG remained associated with ischemic stroke in patients with AF over 70 years old, in patients with AF without a history of smoking, and in patients with AF with CHA₂DS₂-VASc score ≥ 2. Therefore, patients with AF and high TG should consider methods to lower their TG levels, especially if they are over 70 years old, have not smoked, or have CHA₂DS₂-VASc score ≥ 2.

The relationship between blood lipid levels and the risk of ischemic stroke in patients with AF may be complex. It is generally established that the risk of atherosclerosis increases with higher LDL-C [19]. In general populations at risk of ischemic stroke, statin therapy trials have shown that lowering LDL-C by 1 mmol/L reduces the risk of both ischemic stroke and coronary heart disease by about 20% [20]. On the other hand, observational studies have found stronger effects of LDL-C on coronary heart disease than ischemic stroke [21]. These differences highlight the heterogeneity in the effects of cholesterol on different subtypes of ischemic stroke [22]. In Chinese populations, the LDL-C level tends to be lower than in Western populations. Nevertheless, lowering LDL-C in Chinese populations also showed a benefit in preventing ischemic stroke, even below levels considered a risk in Western populations [23]. In China, a previous study suggested a high level of LDL-C was an independent risk factor for stroke in patients with AF [9]. While another study found that reducing oxidized low-density lipoprotein levels with statins prior to stroke reduced and improved outcomes after stroke in patients with AF [10]. However, this study did not find an independent relationship between high LDL-C level and ischemic stroke risk. This result may be influenced by the results of a study that showed patients with AF have reduced LDL-C levels compared to healthy controls [7].

Low HDL-C was reportedly associated with increased possibility the ischemic stroke [24, 25]. However, most recent meta-analysis of the modifiable and non-modifiable risk factors in the Asian population showed that HDL-C was not significantly associated with ischemic stroke [26], while in the review by Kloska et al. decreased HDL-C have been identified as risk factor and predictor of cardiovascular disease, including stroke, in Caucasian population [27]. Conversely, study by Yang found that higher level of HDL was independent indicators of newly-detected AF [28], while Lee et al., also in the Asian population, reported that high TG and low HDL levels were associated with CHD risk (including risk of ischemic stroke) only in participants with an LDL-C level of ≥ 130 mg/dL, but this was not observed in those participants with lower LDL-C levels [29]. The results of our study suggest that the elevation of HDL-C might increase the possibility of the ischemic stroke in patients with AF. This result needs further confirmation, as it may be influenced by non-categorized HDL-C and a lower portion of NOAC users in the stroke group. This difference might lead to some bias. On the one hand, this may be because NOAC is expensive and patients with stroke already incur healthcare costs. On the other hand, perhaps fewer patients use NOAC due to non-compliance, resulting in an uneven distribution. Since the different distribution of NOAC users could introduce bias, NOAC was included as a covariate in the conditional logistic regression models. Still, regarding NOAC therapy, our findings are in line with previous study, reporting decreased ischemic stroke, remarkably low rate of systemic embolism and bleeding in AF patients [30], which suggests that patients receiving NOAC might have lower rate of ischemic stroke. The results of this study did suggest that TG levels were an important factor in ischemic stroke in patients with AF. High TG levels increase the chance of cardiovascular events due to increased atherosclerosis [31]. TG may also stimulate atherogenesis through availability of excessive free fatty acids, production of proinflammatory cytokines, fibrinogen, coagulation factors and impairment of fibrinolysis [32]. Some studies suggest a direct relationship between TG and stroke [33]. Elevated TG caused in relation to genetic factors may not affect stroke risk [34]. Paradoxically, even low TG levels have been shown to play a role in ischemic stroke. Jain et al. studied patients with acute ischemic stroke and found that low TG level was associated with a worse prognosis [35]. Ryu et al. reported that low serum TG is an independent predictor of mortality after ischemic stroke by non-cardioembolic causes [36]. To date, however, the relationship between TG and ischemic stroke in patients with AF has not been fully established. Thus, the results of this study suggest the need to further investigate the relationship between TG and ischemic stroke in patients with AF. TG-lowering methods including diet and lifestyle changes might be considered. When subgroup analysis was applied, the effect of high TG on ischemic stroke risk was maintained in patients over 70 years old but not in the younger group, those without a history of smoking, and those with CHA₂DS₂-VASc score ≥ 2. Therefore, patients in these particular subgroups may find that efforts to reduce TG levels yield the most benefit in terms of reduced incidence of stroke.

This study has some limitations. As this study was performed in a Chinese population, its results cannot entirely be generalized outside of this context. Further studies should be performed with patients of a different ethnicity. Of the total of 25,512 patients with AF enrolled in the China-AF study, only 3839 patients were analyzed and only 722 patients were analyzed in both the groups. Therefore, the potential for selection bias must be considered as a limitation. Although the groups were matched, there were some differences between them in terms of the BMI, DBP, and NOAC use, which may have introduced some bias into the results. While TG was identified as being associated with ischemic stroke, we did not identify a clinically meaningful cutoff value and the effect of decreasing TG levels was not investigated. Further prospective, large-scale randomized clinical trials are needed to further provide high-level evidence.

This study showed that patients with AF who experienced ischemic stroke had higher TG levels compared to matched patients with AF without stroke; however, the TC, LDL-C, and HDL-C levels were similar between both patient groups. Multivariate analysis showed that TG, HDL-C and NOAC were associated with ischemic stroke in patients with AF. TG was also found in subgroup analysis of patients with AF for those aged over 70 years, without a history of smoking, and with CHA₂DS₂-VASc score ≥ 2. These results suggest that studies to investigate the benefit to patients with AF in decreasing TG should be undertaken.