Infrainguinal Lesion of Peripheral Artery Disease and Levels of ω-3 Polyunsaturated Fatty Acids in Peripheral Artery Disease

Abstract

Background: Recently, reduced serum levels of ω-3 polyunsaturated fatty acids (PUFAs) including docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA), have been focused upon as newly recognized risk factors for peripheral artery disease (PAD). The present study investigated the association between disease location and serum levels of ω-3 PUFAs in patients with PAD.

Methods: We retrospectively analyzed the data of patients at Tokyo Medical University between August 2011 and November 2015. The subjects included 98 patients who were categorized into two groups: those with (n=72) and without infrainguinal lesions (n=26).

Results: Univariate analysis revealed that low ankle–brachial pressure index (ABI) values, low EPA levels, low DHA levels, low triglyceride levels, and diabetes mellitus were significant risk factors for infrainguinal lesions. Multivariate analysis indicated that low ABI values [p=0.018; odds ratio, 0.043; 95% confidence interval (CI), 0.003–0.579] and low DHA levels (p=0.003; odds ratio, 0.986; 95%CI, 0.977–0.995) were significant independent risk factors for infrainguinal lesions.

Conclusion: Our study demonstrated that reduced serum level of DHA may underlie the presence of infrainguinal lesions in patients with PAD.

Introduction

The overall incidence of peripheral artery disease (PAD) is 2.7%–7.9% in Japan, and the number of patients has been increasing in the high-age society.¹⁾ PAD also tends to occur more frequently in conjunction with diabetes mellitus (DM) and dyslipidemia. To date, various risk factors for symptomatic or asymptomatic PAD have been reported, including male sex, aging, DM, smoking, hypertension, dyslipidemia, hyperhomocysteinemia, Asian/Hispanic/Black races, C-reactive protein (CRP) level, and renal insufficiency.²⁾ Aging is the most important non-modifiable risk factor and increases the risk of PAD by 2- to 3-fold, whereas the most important modifiable risk factors include smoking and DM, which increase the risk of PAD by 3- to 4-fold.

The locations of PAD vary as follows: femoropopliteal artery in 50%, aortoiliac artery in 24%, tibiofibular artery in 17%, and popliteal artery in 5% of patients.³⁾ These variations are divided roughly into three major patterns of arterial obstruction: proximal lesions (suprainguinal lesions), distal lesions (infrainguinal lesions), and a combination of lesions.^4,5) Patients with proximal lesions more commonly present with intermittent claudication, whereas those with distal lesions frequently have limb-threatening ischemia. Furthermore, several studies have suggested that the etiology of the disease process in the lower extremities may vary depending on the anatomic site of the lesion.⁶⁾ Different clinical risk factors, especially atherosclerotic risk factors, predict the involvement of different arterial segments.

Recently, the reduced levels of plasma ω-3 polyunsaturated fatty acids (PUFAs), such as eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), have been established as newly recognized risk factors for PAD.^7–10) ω-3 PUFAs—found in fish oils—are known to decrease the production of inflammatory eicosanoids, cytokines, and reactive oxygen species and reduce the expression of adhesion molecules.¹¹⁾ Therefore, ω-3 PUFAs have various beneficial effects in lowering the blood pressure and heart rate, improving dyslipidemia, reducing inflammation, and improving vascular and platelet function. Although numerous studies have assessed the role of ω-3 PUFAs in patients with PAD,^7–10) none have investigated the association of ω-3 PUFAs with the disease location in PAD.

We therefore hypothesized that ω-3 PUFAs, especially EPA and DHA, are associated with disease location in PAD and investigated the association between the disease location and coexisting risk factors, including ω-3 and ω-6 PUFAs, in patients with PAD.

Patients and Methods

Patients

This study was approved by the institutional ethics committee of Tokyo Medical University and performed from August 2011 to November 2015, in accordance with the Declaration of Helsinki (2001). The subjects included 98 consecutive patients with PAD who underwent or were scheduled to undergo endovascular therapy (EVT) and/or bypass surgery at Tokyo Medical University Hospital; imaging studies were performed to identify the disease location only in these patients. Exclusion criteria involved patients who denied written informed consent, patients who had undergone previous revascularization for PAD, and patients who were being treated with oral ω-3 PUFA.

The disease location was evaluated using three-dimensional computed tomography angiography, digital subtraction angiography, or duplex scanning. These patients were categorized into two groups: 72 patients (73.5%) with infrainguinal lesions (Group A) and 26 patients (26.5%) without infrainguinal lesions (Group B).

Baseline analysis

The study participants completed questionnaires regarding their health status, medical history, and cardiovascular risk factors and underwent clinical examination. Clinical predictors, including age, male sex, body mass index (BMI), critical limb ischemia (CLI), smoking history, current smoking habits, hypertension, dyslipidemia, DM, history of cerebrovascular disease (CVD), history of ischemic heart disease (IHD), and the presence of end-stage renal failure managed with hemodialysis (HD), were included in the analyses. CLI was defined as Fontaine stages III (pain at rest) and IV (ulcers and gangrene). Hypertension was defined as a systolic/diastolic blood pressure of 140/90 mmHg or if a patient was administered hypertensive medication. DM was diagnosed if a patient had hemoglobin A_1C level of 6.5%, fasting plasma glucose concentration of 126 mg/dL, history of taking any antihyperglycemic medications, or previous diagnosis of diabetes. The ankle–brachial pressure index (ABI) was evaluated using the ABI Form (Omron-Colin, Tokyo, Japan) at our vascular laboratory.

Measurement of serum lipids

To assess lipid profiles, blood samples were obtained from peripheral veins. Serum levels of ω-3 PUFAs, including EPA, DHA, arachidonic acid (AA), and dihomo-γ-linolenic acid (DHLA), were measured at SRL Inc., Tokyo, Japan, using a GC-2010 gas chromatograph (Shimadzu Corporation, Kyoto, Japan) and a TC-70 column (GL Sciences B.V., Tokyo, Japan). Serum levels of lipids, including total cholesterol (TC), triglycerides (TGs), high-density lipoprotein (HDL), and low-density lipoprotein (LDL), were assessed using standardized enzymatic methods at our hospital.

Statistical analysis

All statistical analyses were performed using SPSS 22 (IBM Corporation, Armonk, New York, USA). Continuous values are expressed as mean±standard deviation, and categorical data are expressed as numbers (n) and percentages in parentheses. For continuous data, the normality of the distribution was examined using the Kolmogorov–Smirnov test. Univariate analysis was performed using the Student t-test for normally distributed data and Mann–Whitney U test or chi-square test for other data. Statistically significant variables in univariate analyses were entered into the multivariate model to assess their independent impact on the disease location. In selecting variables, the forward selection method was applied according to the likelihood ratio. The Hosmer–Lemeshow test was used to determine how well the multiple logistic regression model fit our data.

Results

Clinical characteristics of the subjects

The patient characteristics are shown in Table 1. Notably, the serum levels of ω-3 PUFAs also are shown in Table 1: DHA, 139.2±53.3 µg/mL; EPA, 70.5±41.4 µg/mL; AA, 174.2±44.8 µg/mL; and DHLA, 36.0±13.8 µg/mL.

Table 1 Patient characteristics.

Variable
Number of patients	98
Age, years	73±7.3
Male sex	79 (81%)
Infrainguinal lesion	72 (73.5%)
CLI	26 (26.5%)
ABI	0.56±0.21
BMI, kg/m2	22.4±3.3
Smoking	72 (73.5%)
Comorbidities
Hypertension	72 (73.5%)
DM	43 (43.9%)
CVD	31 (31.6%)
IHD	36 (36.7%)
HD	13 (13.3%)
ω-3 fatty acids
EPA, µg/mL	70.5±41.4
DHA, µg/mL	139.2±53.3
AA, µg/mL	174.2±44.8
DHLA, µg/mL	36.0±13.8
Serum lipids
TC, mg/dL	176.4±36.2
TG, mg/dL	151.7±85.9
HDL, mg/dL	46.36±13.9
LDL, mg/dL	99.81±33.5

The values are presented as n (%) or mean±standard deviation, as appropriate. CLI: critical limb ischemia; ABI: ankle–brachial pressure index; BMI: body mass index; DM: diabetes mellitus; CVD: cerebrovascular disease; IHD: ischemic heart disease; HD: hemodialysis; EPA: eicosapentaenoic acid; DHA: docosahexaenoic acid; AA: arachidonic acid; DHLA: dihomo-γ-linolenic acid; TC: total cholesterol; TG: triglyceride; HDL: high-density lipoprotein; LDL: low-density lipoprotein

Univariate analyses concerning disease locations

Table 2 shows comparisons of the clinical characteristics and serum lipid levels between Groups A and B. There were no differences in the values or incidences of age, male sex, BMI, smoking, hypertension, CVD, IHD, and HD, whereas the incidence of DM in Group A was significantly higher than that in Group B (p=0.046). The value of ABI was significantly lower in Group A than in Group B (p=0.046). The serum levels of EPA, DHA, and TG in Group A were significantly lower than those in Group B (p=0.038, 0.008, and 0.048, respectively), but there were no differences in the serum levels of DHLA, AA, TC, HDL, and LDL.

Table 2 Univariate analyses concerning disease location.

Variable	Group A (n=72)	Group B (n=26)	p value
Age, years	73.4±7.3	72.0±7.4	0.398
Male sex, %	79.2	84.6	0.547
ABI	0.54±0.21	0.64±0.20	0.046*
BMI, kg/m2	22.0±3.4	23.4±2.9	0.072
Smoking, %	75.0	69.2	0.568
Comorbidities
Hypertension, %	72.2	76.9	0.642
DM, %	50.0	26.9	0.046*
CVD, %	34.7	23.1	0.274
IHD, %	36.1	38.5	0.831
HD, %	16.7	3.8	0.099
Serum ω-3 fatty acids
EPA, µg/mL	65.1±38.6	85.3±46.6	0.038*
DHA, µg/mL	130.1±47.6	164.4±61.8	0.008*
AA, µg/mL	173.2±46.1	176.8±42.7	0.733
DHLA, µg/mL	35.8±14.1	36.5±13.7	0.832
Serum lipids
TC, mg/dL	175.0±37.6	180.2±33.2	0.531
TG, mg/dL	140.8±84.4	182.0±86.1	0.048*
HDL, mg/dL	46.9±13.6	44.9±15.2	0.523
LDL, mg/dL	99.7±35.8	100.0±27.5	0.968

* p value significant. Abbreviations as in Table 1.

Multivariate analyses concerning disease locations

Multiple regression analysis using significant factors detected in the univariate analyses was performed to identify determinants of the presence of infrainguinal lesions. Consequently, ABI values [p=0.018; odds ratio, 0.043; 95% confidence interval (CI), 0.003–0.579] and DHA levels (p=0.003; odds ratio, 0.986; 95%CI, 0.977–0.995) were found to be significant factors associated with the presence of infrainguinal lesions (Table 3). In contrast, age, male sex, DM, and TG levels did not show any significant correlations with the presence of infrainguinal lesions. The p value for Hosmer–Lemeshow statistics was 0.085 (>0.05), implying that our logistic model fit the data well.

Table 3 Multivariate analyses concerning disease location.

Variable	p value	Odds ratio	95%CI
ABI	0.018*	0.043*1	0.003–0.579
DHA	0.003*	0.986*1	0.977–0.995

* p value significant. *¹ Ratio when a variable changes by 1. Abbreviations as in Table 1.

Discussion

The reasons why atherosclerosis occurs at certain sites, such as the aortoiliac, femoropopliteal, and orbital arteries, remain unclear. Hypotheses explaining the site selectivity of atherosclerotic lesions include hemodynamic stress related to arterial geometry and anatomic, cellular, or biochemical variations in the arterial wall.^6,12) Chen et al. enumerated several specific factors such as the iliac arteries are characterized by elastic properties, whereas the femoropopliteal and tibial arteries contain progressively more muscular elements^6,13); furthermore, the arterial lumen-to-wall thickness decreases from proximal to distal segments, generating both an alteration in arterial flow and shear stress associated with atherosclerosis.^6,14) Therefore, the involvement of different arterial segments may be affected by different clinical factors.

Several studies have demonstrated that the distribution and extent of PAD are affected by atherosclerotic risk factors.⁶⁾ Most recently, Diehm et al. analyzed a consecutive series of 2,659 patients with PAD undergoing EVT and reported that smoking is closely related to proximal disease, whereas higher age and DM are predisposing factors to the involvement of distal disease.¹⁴⁾ Moreover, a striking difference is evident, with more femoropopliteal target lesions detected in female patients and more iliac and infragenicular target lesions in male patients.

ω-3 PUFAs have been focused upon as new atherosclerotic risk factors. It has been reported that reduced plasma levels of ω-3 PUFAs are involved in the deterioration observed in IHD. Low serum EPA levels and low serum EPA to serum AA (EPA/AA) ratios are associated with high vulnerability for coronary plaques.¹⁵⁾ Atherosclerotic plaques readily incorporate EPA, and a higher plaque EPA content is associated with decreased numbers of foam cells and T cells, less inflammation, and increased stability.¹⁶⁾ Until recently, these favorable effects have been primarily attributed to EPA, which is present in large amounts in fish oil. However, several studies now demonstrate that DHA, although often present in lower quantities, has equally important anti-arrhythmic, anti-thrombotic, and anti-atherogenic effects.¹⁷⁾

Several clinical studies have indicated a significant association between serum levels of EPA or ratios of EPA/AA and the incidence of major adverse cardiac events after patients with acute myocardial ischemia underwent percutaneous coronary intervention.^18–20) Although little has been reported regarding DHA, higher plasma DHA levels are associated with the reduced progression of coronary atherosclerosis in women with IHD.²¹⁾ Moreover, the intake of EPA and/or DHA in fish oil has been proven to attenuate the risk of CVD, IHD, and sudden cardiac death in patients with hypercholesterolemia, peripheral artery disease, or atrial fibrillation.^22–24)

More recently, decreased plasma levels of ω-3 PUFAs, including EPA and DHA, have been established as newly recognized factors for PAD. Fujihara et al. demonstrated that reduced plasma EPA/AA ratios may underlie PAD in Japanese patients. ABI is an independently associated factor for PAD.⁷⁾ Fukuda et al. reported that patients with PAD were more likely to have a low EPA/AA ratio, and non-diabetic patients with PAD had a significantly reduced EPA/AA ratio.⁸⁾ Sugiura et al. reported that significant relationships among EPA, CRP, and PAD were confirmed in patients with coronary artery disease (CAD). High CRP levels and low EPA levels were significant and independent predictors of PAD.⁹⁾ Leng et al. revealed that when subjects with PAD were compared with healthy control subjects, the largest differences were noted in plasma levels of ω-3 PUFAs, particularly DHA.¹⁰⁾

This study demonstrates that statistically significant associations exist between serum levels of DHA (but not EPA) and the patterns of disease location in patients with PAD. DHA (p=0.003; odds ratio, 0.986; 95%CI, 0.977–0.995) was found to be a significant risk factor for the presence of infrainguinal lesions. This result may be explained by the differing hemodynamic and anti-atherogenic properties of EPA and DHA.¹⁷⁾ DHA is more effective than EPA in reducing blood pressure, and these blood pressure-lowering effects correlate with improvements in endothelial relaxation and attenuated vascular constriction.²⁵⁾ Platelet aggregatory responses ex vivo and platelet-derived thromboxane A₂ production are reduced by DHA and not EPA.²⁶⁾ Therefore, reduced serum DHA levels may result in greater vascular constriction and platelet aggregation, leading to progression to smaller arteries, such as the infrainguinal arteries.

In this study, ABI was also identified as a significant factor associated with the presence of infrainguinal lesions, and this finding is in agreement with a previous report; ABI values were slightly higher in aortoiliac lesions than in femoropopliteal lesions.²⁷⁾

This study has several limitations. First, the data were retrospectively analyzed from a prospectively maintained database of only patients undergoing EVT. Therefore, there may be some bias in relation to the selection of patients. Second, the cross-sectional nature of the study design may not have allowed the evaluation of the direct effect of DHA on the progression of PAD. Third, the drugs and supplements were assessed but dietary habits were not assessed in this study. This may have affected the outcome of the study.

Conclusion

This study demonstrates that decreased serum levels of DHA and not EPA may underlie the presence of infrainguinal lesions in patients with PAD. Further prospective cohort studies are required to confirm this result.

Disclosure Statement

The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article.

Author Contributions

Conception and design: TN, TI, HO

Analysis and interpretation: TN, TI, HO

Data collection: TN, TI

Writing the article: TN, TI

Critical revision of the article: TN, TI, HO

Statistical analysis: YO, SI

Final approval of the version: TI, TN, YO, SI, HO

Obtained funding: not applicable

Overall responsibility: TN