Associations of Serum Fatty Acids with Serum Lipids Parameters in Ischaemic Stroke Patients

Background: Serum fatty acids (s-FA) are likely associated with serum lipids parameters (s-LP), such as total cholesterol (T-CHO), LDL-cholesterol (LDL-C), HDL-cholesterol (HDL-C) and triglycerides (TG), because fatty acids (FA) are components of lipoproteins. However, little is known about the association of s-FA with s-LP. The aim of our retrospective study was to investigate the association of s-FA with s-LP in Ischaemic stroke patients. Methods: We conducted a cross-sectional study of ischaemic stroke patients aged 50 years or older who 1) were admitted to our institution between September 2015 and March 2017 within 24 hours of stroke onset and 2) took blood examination of s-FA levels of palmitic acid (PA), stearic acid (StA), oleic acid (OlA), linoleic acid (LiA), arachidonic acid (AA), eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). We evaluated correlation of serum levels or composition percentage (%) of FA with s-LP. Results: One hundred ninety-one patients met our inclusive criteria. Their average age was 74.4 years, BMI was 23.4 kg/m2, T-CHO was 203.4, LDL-C was 121.7, HDL-C was 58.5 and TG was 116.2 mg/dl. Stepwise linear regression analysis showed that LiA and AA were independent positive factors of T-CHO, AA and LiA% were independent positive factors of LDL-C, AA was an independent positive factor of HDL-C, AA% and LiA% were independent inverse factors of HDL-C, and OlA was an independent positive factor of TG. Conclusions: Serum FA had correlation with s-LP in ischaemic stroke patients. The results warrant clinical application of s-FA modication to improve s-LP.

Previous studies have reported that n-3 PUFA may reduce the incidence of coronary heart disease or stroke or the mortality associated with cardiovascular disease [3][4][5][6][7]. By contrast, SFA may increase the risks of these conditions [8], and the effects of n-6 PUFA remain controversial [9]. Previous studied have reported that serum n-6 PUFA was inversely and n-9 MUFA positively correlated with TG [10] and linoleic acid lowers LDL cholesterol without a proportionate displacement of SFA [11]. Serum FA may be associated with the age at ischaemic disease onset [12]. However, little is known about the association of s-FA with s-LP particularly in ischaemic stroke patients. The aim of our retrospective study was to investigate the association of s-FA with s-LP in ischaemic stroke patients.

Methods
We conducted a cross-sectional study of ischaemic stroke patients aged 50 years or older who 1) were admitted to our institution between September 2015 and March 2017 within 24 hours of stroke onset and then 2) underwent blood evaluations of s-LP and s-FA. We excluded patients who used statins, n-3 PUFA supplements or puri ed EPA drugs, brates or ezetimibe at onset. Additionally, we excluded those with a pre-hospital modi ed Rankin scale (mRS) score of 3 or more and/or a body mass index (BMI) less than 18.5, which was de ned as severe disability and/or underweight according to the World Health Organisation (WHO) guidelines, as we determined that possible malnutrition was not appropriate to our investigation.

Variables
The

Evaluation
We evaluated correlation among s-FA, correlation between serum concentration levels or composition percentage (%) of FA and s-LP levels.

Statistical analysis
Continuous variables are expressed as means ± standard deviations (SDs) and were analysed by using simple linear regression analysis and stepwise linear regression analysis. A p value of less than 0.05 was considered statistically signi cant. We used the JMP software program (version 14.3; SAS Institute, Cary, NC, USA) to perform the statistical analyses.

Results
A total of 463 patients with ischaemic stroke were admitted to our stroke centre during the study period.
Among them, 129 patients were excluded from our analysis because of a pre-hospital mRS score of 3 or more, 37 were excluded with a BMI of less than 18.5, 57 were excluded because they did not undergo a serum fatty acid analysis at admission and 49 were excluded for using medication for dyslipidaemia. Finally, 191 patients (Table 1) (Table 1). Among fatty acids, LiA(18:2 n-6PUFA) had the largest concentration of 810.6 µg/ml and the largest composition of 27.1% and PA(16:0 SFA) had the second concentration of 711.1 µg/ml, the second composition of 23.7% and OlA(18:1 n-9 MUFA) had the third concentration of 641.2 µg/ml and the third composition of 21.3% (Table 1). The total composition of LiA, PA and OlA was 72.1% among s-FA and the three fatty acids occupied most of s-FA in our patients. On the other hand, the composition of AlA (18:3 n-3), EPA (20:5 n-3) and DHA (22:5 n-3) was 0.8, 2.5 and 4.7% (Table 1), respectively. The total composition of the three n-3 PUFA was 8%. Simple linear regression analysis to explain relationship between fatty acids and serum lipids parameters showed that AA had the largest adjusted R2 to estimate T-CHO level, LDL-C level and HDL-C level (Table 2), although the adjusted R2 was less than 0.5. OlA had the largest adjusted R2 of 0.69 and PA had the second adjusted R2 (0.62) to estimate TG level ( Table 2) (Table 3).
Stepwise linear regression analysis to nd independent determinants of T-CHO showed that the adjusted R2 was 0.49 and that LiA (18:2 n-6 PUFA) ( Figure) and AA (20:4 n-6 PUFA) were positive determinants and OlA% was a negative determinant (Table 4-1). Stepwise linear regression analysis to nd independent determinants of LDL-C showed that the adjusted R2 was 0.21 and that AA (20:4 n-6 PUFA) and LiA% were positive determinants (Table 4-2). Stepwise linear regression analysis to nd independent determinants of HDL-C showed that the adjusted R2 was 0.43 and that AA (20:4 n-6 PUFA) was a positive determinant and AA% and LiA% were negative determinants (Table 4-3). Stepwise linear regression analysis to nd independent determinants of TG showed that the adjusted R2 was 0.72 and that OlA (18:1 n-9 MUFA), AlA (20:4 n-6 PUFA) and PA% were positive determinants (Table 4- with AA (20:4 n-6 PUFA), StA (18:0 SFA) had a stronger correlation with AA (20:4 n-6 PUFA), suggesting that simultaneous ingestion of both StA (18:0 SFA) and AA (20:4 n-6 PUFA) in diet is superior to metabolic pathway from LiA (18:2 n-6 PUFA) to AA (20:4 n-6 PUFA). Vegetable oils, such as soybean oil, usual sun ower oil, grape seed oil and corn oil, contain large amounts of LiA but little of AA and, for example, pork smoked liver contains moderate amounts of StA and AA [14]. Pork smoked liver is cooked with vegetable oils frequently and then larger amounts of LiA and smaller amounts of StA and AA may be ingested simultaneously. If such food is eaten, too large amounts of LiA (18:2 n-6 PUFA) is not correlated with AA (20:4 n-6 PUFA). The concentration of AlA (18:3 n-3 PUFA) was low and AlA (18:3 n-3 PUFA) had weak correlation with EPA (20:5 n-3 PUFA) and no correlation with DHA (22:5 n-3 PUFA), whereas EPA (20:5 n-3 PUFA) had a strong correlation with DHA (22:5 n-3 PUFA). This suggests that metabolic pathway from AlA (18:3 n-3 PUFA) to EPA (20:5 n-3 PUFA) is weak and metabolic pathway from EPA (20:5 n-3 PUFA) to DHA (22:5 n-3 PUFA) may be stronger or both of EPA and DHA may be ingested simultaneously in diet. For example, raw sh like sardine, mackerel and Japanese horse mackerel contains high concentration of EPA and DHA but does not contain AlA [14].
Statins are commonly used to treat hypercholesterolemia and they can reduce serum T-CHO and LDL-C levels [15]. However, a previous study reported that stains were not able to change s-FA levels [16]. In order to prevent progress of atherosclerosis and ischaemic disease, not only statins but also modi cation of s-FA may be required. In our patients, the concentration and composition of LiA(18:2 n-6) was the largest and this may be associated with ischaemic stroke. To modify s-FA, it may be effective to reduce LiA% and ingestion of LiA, to increase OlA and OlA% in diet [17] or to increase AlA and AlA% in diet [18]. A previous study reported that, in individuals with peripheral arterial disease, insu cient evidence exists to suggest a bene cial effect of n-3 PUFA supplementation with regard to major adverse cardiac events, need for revascularization or amputation, pain-free walking disease, or quality of life [19]. Not only supplementation of n-3 PUFA but also appropriate daily intake of fatty acids from dietary sources may prevent abnormal increases in the serum levels of TCHO or TG and would thus inhibit the progression of atherosclerosis. Therefore, further prospective study is needed to determine the appropriate levels of SFA (PA and StA), n-9 MUFA (OlA) and n-6 PUFA (LiA and AA) and su cient levels of n-3 PUFA (EPA and DHA), especially EPA.

Study limitations
Our study had several limitations of note. First, this study included a small number of patients. Moreover, all patients were of East Asian ethnicity and were most likely of Japanese ancestry. In addition to genetic differences attributable to race, our patients likely had different dietary intakes from populations in Western countries. These factors may limit the generalisability of our results. Second, our study design was retrospective and cross-sectional, rather than prospective. Therefore, the appropriate dietary intakes of SFA, MUFA, n-6 PUFA and n-3 PUFA (especially EPA) and consequent serum levels that protect against arteriosclerotic progression could not be con rmed.

Conclusions
Serum fatty acids had correlation with serum lipids parameters in ischaemic stroke patients. The results warrant further clinical application of serum fatty acids modi cation to improve serum lipids parameters.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.