Fish Oil Increases Specialized Pro-resolving Lipid Mediators in PAD (The OMEGA-PAD II Trial)

Abstract

Background:

N-3 polyunsaturated fatty acid (PUFA) supplementation has been associated with reduced mortality and inflammation in patients with cardiovascular disease. There are limited data on the effects of n-3 PUFA supplementation in patients with peripheral artery disease (PAD).

Materials and methods:

The OMEGA-PAD II trial was a double-blinded, randomized, placebo-controlled trial to assess the effect of 3 mo of high-dose oral n-3 PUFA supplementation on inflammation, endothelial function, and walking ability in patients with PAD.

Results:

Twenty-four patients with claudication received 4.4 g/d of fish oil or placebo for 3 mo. Outcomes measured included high-sensitivity C-reactive protein levels, the omega-3 index, endothelial function as measured via flow-mediated vasodilation, walking impairment questionnaire, and a 6-min walk test. Plasma levels of specialized pro-resolving lipid mediators (SPMs) were measured by liquid-chromatography-tandem mass spectrometry. In patients treated with fish oil, the absolute mean omega-3 index significantly increased from baseline (fish oil: 7.2 ± 1.2%, P < 0.001; placebo: −0.4 ± 0.9%, P = 0.31; between-group P < 0.001). Furthermore, there were significant increases in several pathway markers of SPM biosynthesis, including several mono-hydroxyeicosapentaenoic acids and mono-hydroxydocosahexaenoic acids. We also observed significant increases in the SPM lipoxin A₅ (fish oil: 0.57 ± 0.70 pg/mL, P = 0.05; placebo: 0.01 ± 0.38 pg/mL, P = 0.93; between-group P = 0.04) and resolvin E3 (fish oil: 154 ± 171 pg/mL, P = 0.04; placebo: 32 ± 54 pg/mL, P = 0.08; between-group P = 0.04). There were no significant changes in high-sensitivity C-reactive protein, flow-mediated vasodilation, walking impairment questionnaire, or 6-min walk test in the fish oil group.

Conclusions:

Fish oil increases SPMs in plasma of patients with PAD. Further studies are required to determine whether these early changes translate to clinical improvements in patients with PAD.

Introduction

The effects of n-3 polyunsaturated fatty acid (PUFA) supplementation on systemic inflammation^1,2 and cardiovascular disease (CVD)³ have been described in many populations. In addition, studies have shown that n-3 PUFA supplementation is associated with improvements in endothelial function,^4,5 reduction in platelet aggregation,⁶ and improved atherosclerotic plaque stability.⁷ Eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) are the main PUFAs in fish oil and are substrates for enzymes that give rise to specialized proresolving lipid mediators (SPMs), specifically D-series resolvins, E-series resolvins, maresins, and protectins.⁸ These SPMs have receptor-mediated actions on human leukocytes, platelets, and vascular cells and have been shown to actively promote resolution of vascular inflammation and decrease atherosclerosis and vascular injury in animal models.^8–10

Analysis of National Health and Nutrition Examination Survey data suggests that populations consuming a diet high in n-3 PUFAs may have a decreased prevalence of peripheral artery disease (PAD).¹¹ As such, investigators have examined the relationship between n-3 PUFAs and PAD,¹² but heterogeneity in dosage and intervention length has likely contributed to inconsistent results. To more rigorously investigate oral n-3 PUFA supplementation in patients with PAD, the OMEGA-PAD I trial was designed.¹³ This randomized, double-blinded, placebo-controlled trial demonstrated that 1 mo of high-dose n-3 PUFA (4.4 g/d) supplementation can change the metabololipidomic profile in patients with PAD toward increased production of mediators of resolution.¹⁴

Based on these results, we designed the OMEGA-PAD II trial with the hypothesis that high-dose n-3 PUFA supplementation over 3 mo would lead to changes in systemic inflammation, endothelial function, SPM profile, and claudication symptoms.

Methods

Trial design and participants

The OMEGA-PAD II trial is a randomized, double-blinded, placebo-controlled trial. The study protocol was based on the previously published OMEGA-PAD I trial^13,14 and took place at the San Francisco Veterans Affairs Medical Center (SFVAMC) between 2014 and 2016. Patients aged ≥50 y presenting to the outpatient vascular surgery clinic at the SFVAMC with intermittent claudication (Rutherford I-III) and PAD were recruited to the study. PAD was defined as an ankle-brachial index (ABI) of <0.9, toe pressure <70 mm Hg, or ≥50% stenosis in segments of the aortoiliac, femoral, or tibial arteries on imaging. Patients were excluded from participating if they were taking immunosuppressive medications or steroids, had a severe acute illness (e.g., infection, surgery, critical limb ischemia) within the last 30 d, or had severe hepatic (Child-Pugh ≥ B), renal (creatinine ≥ 2 mg/dL), or nonvascular inflammatory disease.

Participants enrolled in the trial were randomized to one of two groups: fish oil or placebo. Patients were randomized by a block randomization with four subjects per block with a ratio of 1:1 for each block. The randomization was done by research pharmacists who maintained the key until the end of the study.

N-3 PUFA supplementation was achieved with four capsules of ProOmega (Ultimate Omega) taken twice daily (Nordic Naturals, Watsonville, CA), corresponding to a total of 4.4 g/d. Each ProOmega (Ultimate Omega) capsule contains 325 mg of EPA and 225 mg of DHA. The dose of 4 g/d corresponds to the American Heart Association’s recommendations for the treatment of hypertriglyceridemia.¹⁵ The placebo group took the same number of capsules containing soybean (Nordic Naturals) that was designed to appear the same as the treatment capsules. Study subjects attended a baseline visit and underwent a comprehensive vascular physiology assessment and then started on the study drug or placebo for 3 mo. After 3 mo of intervention, they returned for a second visit where the assessment was repeated (Supplemental Fig. 1). Compliance with daily supplementation was addressed using a pill count that was performed at the follow-up visit. Dietary information was not collected throughout the trial; however, participants in both groups were encouraged to engage in healthy dietary habits according to national society guidelines.

The primary end point was a change in plasma high-sensitivity C-reactive protein (hsCRP). Several secondary end points were measured to evaluate a wide range of potential effects that n-3 PUFA supplementation may have. These included changes in other biomarkers of inflammation, SPM profile, omega-3 index, brachial artery flow-mediated vasodilation (FMD), and measures of walking ability. SPMs measured included bioactive products and their biosynthetic pathway markers generated from EPA, DHA, and arachidonic acid (AA) (see below).

Institutional review board approval was granted for this study by the Committee on Human Research at the University of California, San Francisco as well as the SFVAMC Research and Development Office, with all participants giving informed written consent. The study was registered with Clinical-Trials.gov (NCT01979874).

Measurements

Demographics, anthropometries, medical history, and hemodynamic measurements

Basic demographic information was provided by the participant through an intake questionnaire. Information on past medical history and medication use was obtained through the SFVAMC electronic medical record, and common atherosclerotic risk factors such as coronary artery disease, hypertension, hyperlipidemia, diabetes mellitus, and smoking history were recorded. ABIs were measured bilaterally using current guidelines and standards.¹⁶

Renal, lipid, metabolic, and inflammation measurements

Blood samples were collected in a fasting state and assayed the same day per standard methodology (Beckman Coulter Analyzer, Miami, FL) for measurement of creatinine, estimated glomerular filtration rate, albumin, hemoglobin A1C, and lipids. Plasma was isolated from venous blood and assayed for hsCRP the same day as collection per standard methodology (Beckman Coulter Analyzer). Serum was stored at −80° C until assayed for interleukin-6 (IL-6) and soluble intracellular adhesion molecule-1 (ICAM-1) using commercially available enzyme-linked immunosorbent assay (ELISA) kits per standard protocol (R&D Systems Inc, Minneapolis, MN). The typical coefficients of variation for IL-6 and ICAM-1 are 7.4% and 4.6%, respectively. The lower limits of detection are 0.04 pg/mL and 0.10 ng/mL, respectively.

Omega-3 index

The omega-3 index represents the red blood cell (RBC) content of the two major long-chain n-3 fatty acids (FAs), EPA and DHA, and equates to EPA + DHA as a percent of total RBC FAs.¹⁷ The RBCs were isolated from whole venous blood and were assayed for n-3 FAs, n-6 FAs, AA, EPA, and DHA, according to the HS-Omega-3 Index methodology.¹⁸ The typical coefficient of variation for the HS-Omega-3 Index using this procedure is 3%. The average omega-3 index in the United States population is 4.5%, with values ranging from 2.7% in the lowest fifth percentile to 8.8% in the highest 95th percentile.¹⁹

Mass spectrometry—based lipid mediator metabolomics

Frozen plasma samples were subjected to solid-phase extraction and profiled for bioactive lipid mediators using liquid-chromatography-tandem mass spectrometry (LC-MS/ MS) using methodologies that have been previously described.²⁰ Briefly, three volumes of methanol containing internal deuterium-labeled standards (i.e., d5-RvD2, d5-LXA₄, d4-PGE₂, d4-LTB₄, d8–5-HETE) were added to plasma samples before solid-phase extraction to assess extraction recovery in each chromatographic region. Methyl formate fractions were collected and dried under a steady stream of N₂ gas, resuspended in methanol:water (50:50), and analyzed by LC-MS/MS. Identification of mediators was accomplished using specific multiple reaction monitoring transitions and matching of retention time and diagnostic fragmentation spectra as compared to authentic standards. Abundance of lipid mediators was quantified using standard curves constructed with synthetic or authentic standards for each compound.

Brachial artery FMD

Brachial artery FMD was measured according to current guidelines and standards²¹ and as already described.^14,22 FMD in healthy subjects is expected to be above 7%²¹ and has been reported to range between 0.2 and 19.2%.²³

Functional tests

The 6-min walk test was administered according to standard procedures,²⁴ and the distance to claudication and the time to claudication were recorded. Patients completed the walking impairment questionnaire, which is a validated survey that assesses a patient’s perceived walking capacity and limitation due to claudication across three domains: distance, speed, and stair climbing.²⁵

Statistical analysis

Sample size was estimated based on the primary end point (reduction in hsCRP). It was estimated that a hsCRP value of 5.0 ± 5.0 mg/L can be expected in the PAD population¹³ and that 3 mo of n-3 PUFA supplementation would result in a 30% decrease in hsCRP.²⁶ A sample size of 30 patients per group (60 in total) would have 80% power to conclude that hsCRP reduction is significantly higher in the treatment group.

Statistical analyses were performed using STATA 15 (StataCorp, College Station, TX), and variables were summarized by appropriate descriptive statistics. Baseline demographics and clinical variables were compared between the placebo and fish oil group using Fisher’s exact test for categorical variables and Student’s t-test for continuous variables. Paired Student’s t-tests were used to compare baseline variables with postintervention variables. All analyses were based on intention to treat.

To visualize changes driven by n-3 PUFA supplementation, interaction network pathway analyses of the EPA lipid mediator metabolome were performed between treatment groups using Cytoscape. To accomplish these analyses, a missing value imputation that replaced nondetected values with half the minimum value for each mediator was used. In addition, a log transformation of the data was performed. These pathways graphically illustrate both the magnitude of change from baseline to 3-month follow-up for each treatment and the mean abundance of each mediator in the 3-month follow-up samples.

Results

Twenty-four patients were enrolled in the study (all male), with 11 randomized to the fish oil group and 13 randomized to the placebo group (Supplemental Fig. 2). Recruitment for the trial was slower than expected and was ended by the principal investigator before reaching target enrollment. Two subjects (one per group) did not complete a follow-up visit and dropped out of the trial because of personal reasons. One subject in the placebo group developed headaches of an unknown cause and was removed from the study, reported to the institutional review board, and referred to the neurology service, who did not suspect that the placebo pills were the cause of the headaches. One subject in the fish oil group was hospitalized for a stroke and stopped taking the intervention while in the hospital. No participant in this trial dropped out of the study because of gastrointestinal upset, which is a commonly reported adverse effect of fish oil supplementation. This resulted in an overall dropout rate of 17%. Both groups were balanced after randomization with the exception of lower aspirin use in the fish oil (7/11 versus. 13/13, P = 0.03) (Table 1).

Table 1 —

Baseline characteristics of participants

Characteristics	Fish oil (n = 11)	Placebo (n = 13)	P value*
Age (y)	69 ± 8	73 ± 7	0.14
Male sex	11 (100%)	13 (100%)	1.0
Caucasian	7 (63%)	10 (77%)	0.66
BMI (kg/m2)	29 ± 5	28 ± 5	0.56
Waist-hip ratio	0.99 ± 0.05	1.02 ± 0.07	0.32
History of smoking	10 (91%)	13 (100%)	0.46
Pack y	43 ± 35	31 ± 18	0.28
Omega-3 index (%)	5.5 ± 2.1	5.7 ± 1.5	0.77
Index ABI	0.60 ± 0.09	0.66 ± 0.09	0.11
Rutherford
Mild claudication	3 (27%)	5 (39%)	0.86
Moderate claudication	7 (64%)	6 (46%)
Severe claudication	1 (9%)	2 (15%)
History of revascularization	4 (36%)	3 (23%)	0.66
Lower extremity bypass	2 (50%)	1 (33%)	0.86
Lower extremity percutaneous	2 (50%)	2 (67%)	1.0
Comorbidities
Hypertension	9 (82%)	11 (85%)	1.0
Hyperlipidemia	8 (73%)	12 (92%)	0.30
Type 2 diabetes mellitus	3 (27%)	4 (31%)	1.0
Coronary artery disease	4 (36%)	7 (54%)	0.44
Systolic blood pressure (mm Hg)	149 ± 20	154 ± 22	0.52
Diastolic blood pressure (mm Hg)	78 ± 8	82 ± 12	0.35
Medications
Aspirin	7 (64%)	13 (100%)	0.03
ACE-inhibitor	5 (45%)	6 (46%)	1.0
Beta-blocker	6 (55%)	7 (54%)	1.0
Statin	11 (100%)	12 (92%)	1.0
Laboratory studies
Total cholesterol (mg/dL)	153 ± 38	171 ± 52	0.36
LDL (mg/dL)	87 ± 33	79 ± 18	0.53
HDL (mg/dL)	48 ± 11	54 ± 21	0.45
Triglycerides (mg/dL)	92 ± 42	145 ± 116	0.17
Serum Cr, mg/dL	0.91 ± 0.2	1.10 ± 0.3	0.08
eGFR (mL/min)	85 ± 20	71 ± 24	0.14
Albumin (g/dL)	4.1 ± 0.3	4.2 ± 0.3	0.35
HbA1C (%)	5.9 ± 0.7	5.9 ± 0.9	0.99
Vitamin D (ng/mL)	31 ± 10	29 ± 14	0.75
Inflammation
hsCRP (mg/L)	3.4 ± 2.4	3.8 ± 6.2	0.85
IL-6 (pg/mL)	1.4 ± 0.8	2.5 ± 3.4	0.35
ICAM-1 (ng/mL)	576 ± 204	535 ± 141	0.57
Brachial artery FMD
Brachial FMD (%)	8.0 ± 4.6	5.8 ± 4.5	0.25
Patient-perceived walking performance
Walking distance (score, from 0 to 100)	42 ± 32	42 ± 35	0.99
Walking speed (score, from 0 to 100)	38 ± 18	41 ± 30	0.77
Stairs climbing (score, from 0 to 100)	40 ± 34	49 ± 30	0.50
6-min walk test
Time to claudication (s)	113 ± 86	165 ± 99	0.30
Distance to claudication (meters)	119 ± 74	219 ± 125	0.07

Values are as “means ± SD” or “n (%).” Boldface P values were below the 0.05 level required for statistical significance.

ABI = ankle-brachial index; ACE = angiotensin-converting enzyme; BMI = body mass index; bpm = beats per minutes; Cr = creatinine; eGFR = estimated glomerular filtrate rate; FMD = flow-mediated vasodilation; HDL = high-density lipoprotein; HbA1c = hemoglobin A1c; hsCRP = high-sensitivity C-reactive protein; ICAM-1 = intercellular adhesion molecule-1; IL-6 = interleukin-6; LDL = low-density lipoprotein.

^*Calculated using Fisher’s exact test for categorical variables or a two-tailed Student’s t-test for continuous variables.

There was no significant change in the primary outcome of hsCRP levels after intervention in the fish oil or placebo group (Table 2). This was also true for IL-6 and ICAM-1. However, the omega-3 index increased more than two-fold in the fish oil group only (7.2 ± 1.2% increase from baseline, P < 0.001; between group differential P < 0.001).

Table 2 —

Changes in lipid, inflammatory, hemodynamic profile, and vascular function with treatment.

Measurement	Fish oil (n = 11)		Placebo (n = 13)		Difference between groups†
	Change compared to baseline	P value*	Change compared to baseline	P value*
Omega-3 index
Omega-3 index (%)‡	7.2 ± 1.2	<0.001	−0.4 ± 0.9	0.31	<0.001
Lipid profile
Total cholesterol (mg/dL)	−8.7 ± 21	0.26	−5.7 ± 14	0.20	0.71
Triglycerides (mg/dL)	−14 ± 46	0.40	6 ± 28	0.51	0.26
HDL (mg/dL)	2.2 ± 6.2	0.31	−1.5 ± 7.2	0.52	0.25
LDL (mg/dL)	−8.2 ± 25.1	0.35	−3.3 ± 8.6	0.26	0.57
Inflammation
hsCRP (mg/L)	0.8 ± 5.5	0.67	−0.7 ± 3.4	0.52	0.47
IL-6 (pg/mL)	−0.2 ± 0.6	0.37	−0.4 ± 2.4	0.64	0.87
ICAM-1 (ng/mL)	71 ± 145	0.20	16 ± 139	0.70	0.41
Hemodynamic parameters
Systolic blood pressure (mm Hg)	−2.5 ± 12	0.50	−0.5 ± 23	0.93	0.79
Diastolic blood pressure (mm Hg)	−0.8 ± 8.3	0.75	−1.5 ± 10.8	0.62	0.86
Index ABI	0.06 ± 0.18	0.34	0.05 ± 0.11	0.16	0.81
Patient-perceived walking Performance
Walking distance (score, from 0 to 100)	11 ± 14	0.06	1 ± 26	0.87	0.35
Walking speed (score, from 0 to 100)	0.3 ± 23.5	0.97	1.2 ± 31.6	0.90	0.95
Stairs climbing (score, from 0 to 100)	4.2 ± 17	0.47	3.1 ± 18	0.58	0.89
6-min walk test
Time to claudication (s)	94 ± 88	0.12	7 ± 89	0.83	0.14
Distance to claudication (meters)	106 ± 92	0.06	−6 ± 99	0.86	0.06
Brachial artery FMD
Brachial FMD (%)‡	−1.5 ± 3.7	0.27	2.9 ± 3.6	0.03	0.02

Values are as “means ± SD.” Boldface P values were below the 0.05 level required for statistical significance.

ABI = ankle-brachial index; FMD = flow-mediated vasodilation; HDL = high-density lipoprotein; hsCRP = high-sensitivity C-reactive protein; ICAM-1 = intercellular adhesion molecule-1; IL-6 = interleukin-6; LDL = low-density lipoprotein.

^*Calculated using a paired Student’s t-test.

^†Calculated using an unpaired Student’s t-test.

^‡Absolute difference between follow-up and baseline visit.

Consistent with increases in the omega-3 index observed in the fish oil group, several downstream omega-3 PUFA products increased in this group as well. Fish oil supplementation resulted in increases in total plasma DHA (2046 ± 2094 pg/mL, P = 0.03) and EPA (6785 ± 2039 pg/mL, P < 0.001) (Fig. 1). The increase in EPA observed in the fish oil group was significantly greater than the change in EPA observed in the placebo group (P < 0.001). Using interaction network pathway analyses to visualize quantitative differences in the EPA metabolome from baseline to follow-up, we observed enrichment in SPM biosynthesis pathways in the fish oil group (Fig. 2). Several downstream EPA products that are SPM biosynthesis pathway markers were identified (Table 3 and Fig. 2). These included 15-hydroxyeicosapentaenoic acid (15-HEPE), a marker of lipoxin A5 (LXA₅) biosynthesis, and 18-HEPE, which is a marker of E-series resolvin biosynthesis.⁸ The levels of 15-HEPE and 18-HEPE increased significantly after 3 mo of fish oil supplementation, whereas there was no change in the placebo group (Table 3 and Fig. 3). Similarly, levels of LXA₅ increased in the fish oil group but remained the same in the placebo group. E-series resolvins (i.e., RvE1 and RvE3) increased in the fish oil group, but only changes in RvE3 were statistically significant (Table 3 and Fig. 3).

Fig. 1 —

N-3 polyunsaturated fatty acid (PUFA) supplementation (ProOmega) increases levels of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) in the plasma. *Statistical significance as defined by a P value < 0.05. (Color version of figure is available online.)

Fig. 2 —

Interaction network pathway analysis of the eicosapentaenoic acid (EPA) metabolome after n-3 polyunsaturated fatty acid supplementation (ProOmega) or placebo is shown in (A). Representative MS/MS fragmentation spectra used for identification of lipoxin A₅ (LXA₅) and resolvin E3 (RvE3) are shown in (B). (Color version of figure is available online.)

Table 3 —

Changes in DHA- and EPA-derived lipid mediator profiles with n-3 PUFA supplementation.

Product	Fish oil (n = 11)			Placebo (n = 13)			Difference between groups†
	Pre [range]	Post [range]	P value*	Pre [range]	Post [range]	P value*
Plasma EPA	2783 ± 1244 [1528, 5464]	9568 ± 2546 [5829, 13,280]	<0.001	3773 ± 1979 [1182, 6082]	4023 ± 2480 [1559, 8612]	0.58	<0.001
EPA products
5-HEPE	11 ± 11 [2, 37]	67 ± 42 [31, 139]	0.003	10.0 ± 5.9 [3, 23]	14 ± 10 [5, 43]	0.07	<0.001
11-HEPE	3.2 ± 2.7 [1, 10]	39 ± 32 [17, 116]	0.02	4.7 ± 3.2 [1, 10]	5.9 ± 3.8 [2, 14]	0.28	0.002
12-HEPE	24 ± 21 [3, 63]	90 ± 64 [9, 201]	0.02	142 ± 165 [6, 371]	67 ± 42 [14, 170]	0.17	0.04
15-HEPE	19 ± 16 [9, 57]	99 ± 49 [45, 192]	0.002	20 ± 12 [5, 39]	24 ± 13 [5, 43]	0.08	<0.001
18-HEPE	23 ± 16 [12, 62]	195 ± 119 [113, 469]	0.005	36 ± 20 [11, 68]	41 ± 32 [13, 121]	0.47	<0.001
LXA5	0.05 ± 0.13 [0, 0.4]	0.62 ± 0.66 [0, 2]	0.05	0.19 ± 0.31 [0, 1]	0.20 ± 0.56 [0, 2]	0.93	0.04
RvE1	0.32 ± 0.90 [0, 3]	0.69 ± 0.83 [0, 2]	0.06	0.33 ± 0.88 [0, 3]	0.83 ± 1.10 [0, 3]	0.22	0.78
RvE2	13 ± 14 [0, 32]	9.7 ± 3.5 [0, 22]	0.54	13 ± 31 [0, 101]	6.2 ± 6.1 [0, 17]	0.54	0.83
RvE3	33 ± 38 [0, 107]	187 ± 199 [0, 539]	0.04	26 ± 38 [0, 119]	58 ± 82 [0, 183]	0.08	0.04
Plasma DHA	7252 ± 3083 [4265, 12,918]	9298 ± 2166 [5539, 11,951]	0.03	6629 ± 2356 [3542, 10,312]	7121± 2344 [3878, 11,734]	0.23	0.06
DHA products
4-HDHA	3.1 ± 3.0 [1, 10]	11.0 ± 2.8 [8, 16]	0.001	5.0 ± 2.9 [2, 9]	5.6 ± 3.9 [2, 15]	0.48	<0.001
7-HDHA	3.4 ± 2.1 [0, 6]	4.7 ± 1.7 [3, 7]	0.07	2.0 ± 1.2 [1, 4]	2.8 ± 2.0 [1, 6]	0.23	0.59
13-HDHA	10.2 ± 6.5 [5, 26]	28.3 ± 9.2 [20, 49]	0.004	11.0 ± 6.8 [2, 26]	12.1 ± 7.9 [3, 28]	0.55	<0.001
14-HDHA	11 ± 11 [1, 32]	44 ± 27 [20, 86]	0.01	9.0 ± 4.7 [2, 17]	24 ± 24 [3, 77]	0.04	0.12
17-HDHA	25 ± 21 [11, 76]	90 ± 30 [48, 128]	<0.001	27 ± 16 [4, 56]	38 ± 30 [6, 92]	0.10	<0.001
21-HDHA	9.7 ± 6.8 [4, 23]	31.7 ± 8.8 [22, 48]	<0.001	9.1 ± 4.9 [3, 17]	12 ± 12 [3, 44]	0.23	<0.001
RvD1	0.10 ± 0.27 [0, 1]	0.12 ± 0.35 [0, 1]	0.87	0 ± 0 [0, 0]	0.16 ± 0.38 [0, 1]	0.21	0.52
RvD2	0 ± 0 [0, 0]	15 ± 30 [0, 87]	0.19	0.9 ± 1.8 [0, 5]	1.5 ± 1.9 [0, 5]	0.48	0.12
RvD3	0.13 ± 0.17 [0, 0.4]	0.17 ± 0.23 [0, 1]	0.55	0.04 ± 0.07 [0, 0.2]	0.03 ± 0.07 [0, 0.2]	0.92	0.55
RvD4	2.1 ± 0.4 [0, 17]	0 ± 0 [0, 0]	0.35	7.7 ± 20.5 [0, 68]	59 ± 191 [0, 636]	0.34	0.39
RvD5	0 ± 0 [0, 0]	0.58 ± 1.09 [0, 3]	0.18	0 ± 0 [0, 0]	0 ± 0 [0, 0]	NA	0.09
17R-RvD1	0.01 ± 0.04 [0, 0.1]	0.11 ± 0.23 [0, 1]	0.30	0.5 ± 1.7 [0, 6]	0.9 ± 2.6 [0, 9]	0.23	0.43
17R-RvD3	0 ± 0 [0, 0]	0.03 ± 0.05 [0, 0.1]	0.11	0 ± 0 [0, 0]	0 ± 0 [0, 0]	NA	0.05
PD1	2.3 ± 5.0 [0, 15]	5.1 ± 12.6 [0, 36]	0.34	5.6 ± 14.2 [0, 48]	3.6 ± 5.8 [0, 17]	0.61	0.35
17R-PD1	1.8 ± 5.1 [0, 15]	0.9 ± 2.5 [0, 7]	0.35	4.6 ± 7.7 [0, 20]	4.2 ± 9.4 [0, 31]	0.92	0.91
10S, 17S-diHDHA	0.02 ± 0.05 [0, 0.2]	0.31 ± 0.82 [0, 2]	0.36	0.08 ± 0.21 [0, 1]	2.5 ± 6.2 [0, 21]	0.22	0.35
MaR1	0.13 ± 0.20 [0, 0.53]	0.24 ± 0.23 [0, 0.54]	0.20	0.04 ± 0.09 [0, 0.3]	0.06 ± 0.13 [0, 0.4]	0.76	0.31
MaR2	2.2 ± 2.0 [0, 5.4]	2.5 ± 2.5 [0, 7]	0.51	0.9 ± 1.6 [0, 5]	1.3 ± 2.0 [0, 6]	0.10	0.82
7S,14S-diHDHA	0.01 ± 0.03 [0, 0.1]	0.04 ± 0.11 [0, 0.3]	0.35	0.16 ± 0.39 [0, 1]	0.22 ± 0.35 [0, 1]	0.45	0.75

Values are as “means ± SD” in units of pg/mL. Boldface P values were below the 0.05 level required for statistical significance.

7S,14S-diHDHA = 7S,14S-dihydroxydocosahexaenoic acid; DHA = docosahexaenoic acid; EPA = eicosapentaenoic acid; HDHA = hydroxydocosahexaenoic acid; HEPE = hydroxyeicosapentaenoic acid; LXA5 = lipoxin A5; MaR1 (2) = maresin 1 (2); n-3 PUFA = n-3 polyunsaturated fatty acid; PD1 = protectin D1; RvD1 (2, 3, 4, 5) = resolvin D1 (D2, D3, D4, D5); RvE1 (2, 3) = resolvin E1 (E2, E3).

^*Calculated using a paired Student’s t-test.

^†Calculated using an unpaired Student’s t-test.

Fig. 3 —

N-3 polyunsaturated fatty acid (PUFA) supplementation (ProOmega) increases levels of monohydroxy intermediates (A) and specialized pro-resolving lipid mediators (B) generated from eicosapentaenoic acid (EPA). *Statistical significance as defined by a P value < 0.05.. HEPE = hydroxyeicosapentaenoic acid; LXA₅ = lipoxin A₅; RvE1, E2, E3 = resolvin E1, E2, E3. (Color version of figure is available online.

Similar to products of EPA metabolism, several downstream DHA products that serve as SPM biosynthesis pathway markers were identified in the plasma. In the fish oil group, levels of 14-hydroxydocosahexaenoic acid (14-HDHA) and 17-HDHA, which are markers of the maresin and D-series resolvin pathways, respectively, increased significantly after fish oil supplementation (Table 3). We also identified D-series resolvins (RvD1–5), maresins (MaR1, MaR2), and protectins (PD1,17R-PD1) in the plasma (Table 3). However, no significant relationship between their levels and fish oil supplementation was observed. Similarly, we identified several AA products, including lipoxins, leukotrienes, and prostaglandins, but no significant relationships between the fish oil and placebo groups were observed (Supplemental Table 1).

With regard to functional measurements, after intervention, the fish oil group had a greater time to claudication (94 ± 88 s increase from baseline, P = 0.12) and a greater distance to claudication (106 ± 92 m increase from baseline, P = 0.06) after intervention as measured by the 6-min walk test (Table 2). Although these changes were not statistically significant, these trends were not observed in the placebo group. There were no significant changes in perceived walking performance as measured by the walking impairment questionnaire in either group. In addition, there was no significant change in FMD in the fish oil group, although FMD increased in the placebo group, which was significantly different between groups (Table 2).

Discussion

The OMEGA-PAD II trial was designed to investigate the effects of 3 mo of high-dose n-3 PUFA supplementation on inflammation and vascular function in patients with PAD. No difference in hsCRP was observed between the fish oil and placebo group, and no significant differences in functional outcomes were observed. However, the omega-3 index, a predictor of cardiovascular risk,²⁷ increased more than twofold in the fish oil group with no significant changes noted in the placebo group. In addition, there were significant increases in downstream EPA and DHA products, including several SPMs, which have been demonstrated to be potent mediators of inflammation-resolution and have biological roles that could potentially explain several previously reported cardioprotective effects of n-3 PUFAs.^8,9 Although the OMEGA-PAD I trial originally identified significant increases in intermediates of SPM biosynthesis (e.g., HDHA and HEPE) after 1 mo of fish oil supplementation, it did not identify significant changes in SPM end products (e.g., resolvins and lipoxins) as reported in the current trial.

Primary end point

PAD is associated with elevated levels of inflammation, and markers of inflammation have been identified as predictors of mortality²⁸ and poor surgical outcomes.²⁹ N-3 PUFA consumption has previously been reported to be associated with lower serum levels of several inflammatory markers.¹ Siasos et al. reported reductions in IL-6 and tumor necrosis factor-α after 3 mo of oral n-3 PUFA supplementation.²⁶ However, participants in that study did not have preexisting CVD, PAD, or any known clinical atherosclerosis. Although there are limited data analyzing the effects of oral n-3 PUFA supplementation specifically in patients with PAD, Schiano et al. reported no changes in hsCRP levels after 3 mo of n-3 PUFA treatment.³⁰ The current trial did not detect any significant changes in hsCRP, IL-6, or ICAM-1. These gross measures of systemic inflammation might not adequately measure local inflammation at the vascular level and might not be the best way to assess the effects of n-3 PUFA supplementation on atherogenesis and PAD. It is also possible that the small sample size of this trial did not allow for the detection of changes in inflammatory markers or that 3 mo of n-3 PUFA supplementation is an inadequate time period to see changes in these biomarkers.

Secondary end points

Deficiencies in the levels of EPA and DHA in RBCs, as measured by the omega-3 index, have been implicated with adverse cardiac events.²⁷ Efforts to address deficiencies in the omega-3 index could result in reduced morbidity and mortality, which is particularly applicable to PAD cohorts given their numerous atherosclerotic risk factors and high risk for adverse cardiac events.³¹ The OMEGA-PAD I trial reported an absolute mean increase of 4% in the omega-3 index after 1 mo of fish oil supplementation,¹⁴ whereas this trial reports an absolute mean increase of more than 7% after 3 mo of fish oil supplementation.

The mechanism through which n-3 PUFA supplementation and the omega-3 index could have protective effects has been proposed to be via increasing the production of mediators of resolution of inflammation, specifically SPMs. Recent research demonstrates that the resolution of inflammation is an active process driven by SPMs, which are derived from n-3 and n-6 PUFAs.⁸ These SPMs are generated via specific biosynthetic pathways and represent distinct classes including lipoxins derived from AA or EPA; the E-series resolvins generated from EPA; and the DHA-derived D-series resolvins, protectins, and maresins.⁸ SPMs have been shown to have potent proresolution effects in several models of disease, including atherosclerosis.^8,10,32 As SPMs orchestrate termination of inflammation and return to tissue homeostasis, they may be protective against atherosclerosis, vascular injury, and PAD.^33–36 This trial, despite the small sample size, provides substantial evidence that patients with PAD are capable of utilizing EPA and DHA in endogenous enzymatic pathways that increase the production of SPMs.

Lipoxins and resolvins regulate leukocyte-endothelial interactions,^37,38 reduce the formation of reactive oxygen species,³⁹ and regulate the production of prostacyclin⁴⁰ and nitric oxide.⁴¹ Although clinical studies assessing the role of SPMs in PAD are sparse, Ho et al. reported a lower level of 15R-LXA₄ in patients with PAD when compared with healthy subjects,³³ suggesting that patients with PAD might have a deficit of mediators of resolution. A growing body of evidence suggests that mediators of resolution and n-3 PUFA isolates, such as icosapent ethyl, play a protective role in CVD.^42,43 Results of this trial further support that n-3 PUFA supplementation increases plasma levels of vasculoprotective SPMs.

There are several reports of n-3 PUFA supplementation improving endothelial function⁴⁴; however, there are limited data analyzing the effects of supplementation specifically in patients with PAD. Schiano et al. reported an improvement in FMD in patients with PAD after 3 mo of n-3 PUFA supplementation when compared with their control group.³⁰ The OMEGA-PAD I trial reported an improvement in FMD after 1 mo of n-3 PUFA supplementation, but this difference was not significant when compared with changes observed in the placebo group.¹⁴ Results from the OMEGA-PAD II trial report a paradoxical decrease in FMD in the fish oil group and increase in FMD in the placebo group. These results are surprising, and the mechanism underlying them is unclear. Although measurements of FMD were done by trained and experienced staff, there are several factors that can acutely alter the results of FMD measurements including caffeine use, medication use, exercise, anxiety, and stress. Patients were directed to refrain from activities that are known to affect FMD results before measurement, but not all of these factors are controllable and may have contributed to the paradoxical results reported in this study. These results contradict previous evidence and should be further explored using larger sample sizes.

Although not statistically significant, results of this trial suggest that n-3 PUFA supplementation may play a role in increasing time to claudication and walking distance. Previous studies have assessed the effects of n-3 PUFA supplementation on functional outcomes, such as walking distance, and have reported mixed results. Carrero et al. and Madden et al. both examined the effects of n-3 PUFAs in intermittent claudicants and reported improvements in walking distance.^45,46 In contrast, a recent meta-analysis representing a total of 425 participants from nine randomized placebo-controlled clinical trials that measured the effects of n-3 PUFAs on claudication symptoms failed to identify significant improvements between treatment and control groups in pain-free walking distance and maximal walking distance.⁴⁷ However, all of the trials included in this meta-analysis varied in supplementation length, dose, and n-3:n-6 PUFA ratio, which could have contributed to the suboptimal results.

Limitations

This study is limited by a small sample size. Trial recruitment was closed early, and only 40% of the target sample was achieved. The comprehensive vascular physiology assessment that was completed at each study visit amounted to more than 6 h and this time commitment likely impaired study enrollment. In addition, the sample calculation was based on a baseline mean hsCRP of 5.0 mg/L, yet the mean hsCRP of this sample was much lower, and therefore even a sample size of 60 would have been underpowered to detect a difference in the primary end point. The groups were largely balanced after randomization, but differences in aspirin use could have altered the results. In addition, the use of cilostazol was not recorded and may have affected our results. Although not significant, patients in the fish oil group had a higher baseline FMD than the placebo group and could represent falsely elevated baseline FMD, which may have affected the postintervention results. Although recruitment was open to men and women, this cohort reflects the Veterans Affairs population and was made up completely of men, which limits the generalizability of the results.

Conclusion

Although the role of n-3 PUFA supplementation in PAD is poorly understood, there have been limited data that support their beneficial effects in patients with PAD, including the OMEGA-PAD I trial. In this follow-up trial, 3 mo of n-3 PUFA supplementation resulted in significant increases in the omega-3 index and SPMs and their biosynthetic pathway markers in patients with PAD. This provides evidence of the potential role that n-3 PUFA, or SPM, supplementation may play in addressing an inflammation resolution deficit in patients with PAD. Further research utilizing high-dose n-3 PUFA supplementation in larger, more diverse, cohorts is required to confirm and expand the results of this trial.