Recent insights into dietary ω-6 fatty acid health implications using a systematic review

Abstract

The American Heart Association suggests that consuming ω-6 fatty acids (5–10% of total energy) can prevent cardiovascular disease by improving lipoprotein profiles. However, some studies warn of deleterious effects of these due to eicosanoid biosynthesis. We explored the five years for clinical evidence of ω-6 fatty acids on several diseases including inflammation, cancer, cardiovascular disease, and metabolic syndrome. Predefined criteria identified a total of 21 articles in 5 databases. Some studies indicated that dietary arachidonic acid was not related to increase of pro-inflammatory cytokines. In cohort studies, ω-6 fatty acids prevented the onset of digestive and lung cancer. ω-6 Fatty acids improved blood lipoprotein profiles. Moreover, consuming ω-6 fatty acids delayed diabetes mellitus and chronic renal disease and had positive effects on muscle recovery and glaucoma. In conclusion, ω-6 fatty acids have beneficial effects on cancers, blood lipoprotein profiles, diabetes, renal disease, muscle function, and glaucoma without inflammation response.

Introduction

Fatty acids are the primary components of lipids (or fats), in which a methylene group (-CH2-) is linked between a carboxyl end (-COOH) and methyl end (-CH3). Fatty acids are classified as saturated fatty acids (SFA) and unsaturated fatty acids based on presence of a double bond. The position of the first double bond from the methyl end divides fatty acids ω-6 from ω-3. ω-6 fatty acids include linoleic acid (LA, C18:2 n6), γ-linolenic acid (18:3 n6), dihomo-γ-linolenic acid (20:3 n6), and arachidonic acid (AA, C20:4 n6). ω-3 fatty acids include α-linolenic acid (LNA, C18:3 n3), eicosapentaenoic acid (EPA, C20:5 n3), and docosahexaenoic acid (DHA, C22:6 n3). LA and LNA, classified as essential fatty acids, need to be acquired from dietary sources due to lack of endogenous synthesis. The major dietary source of LA is corn oil, soybean oil, and walnuts, while LNA is found in flaxseed oil, chia seeds, and fish oils (Saini and Keum, 2018). Essential fatty acids are precursors to fatty acids synthesized by desaturase and elongase enzymes (Kaur et al., 2014; Naughton et al., 2016). LA is converted to γ-linolenic acid by Δ6 desaturase acid and is converted subsequently to dihomo-γ-linolenic acid by elongase enzymes (Wall et al., 2010). From this, Δ5 desaturase synthesizes AA. LNA is converted to EPA and DHA through Δ5 and Δ6 desaturase and elongase enzyme processes. LA and LNA compete for desaturase and elongase throughout this process, and a large amount of LA intake can hinder LNA conversion to EPA and DHA. Therefore, it is necessary to determine the optimal balance between ω-3 and -6 (Lands, 2014). A low ω-6:ω-3 fatty acid ratio is healthier with respect to inflammation (DiNicolantonio and O'Keefe, 2018), though proper intake research is imperative for understanding the health implications of fatty acid type and intervention duration.

One of the major functions of essential fatty acids is to synthesize eicosanoids, eicosanoids are a complex family containing prostaglandins (PGE and PGI), leukotrienes (LTB), and thromboxanes (TXA and TXB). The physiological effects of eicosanoids differ depending on derivation from ω-3 or ω-6 fatty acid. EPA-derived compounds include PGE3, PGI3, and TXB3 and are likely anti-arrhythmic, anti-inflammatory, platelet inhibitor, and vasodilative (Dennis and Norris, 2015). On the other hand, AA-derived compounds, which include PGE2, PGI2, LTB2, TXA2, and TXB2, are mostly pro-arrhythmic, pro-inflammatory, platelet activator, and vasoconstrictive (Harizi et al., 2008; Wang and Dubois, 2010). For this reason, many studies have investigated health risks associated with ω-6 fatty acids and eicosanoid synthesis. For instance, dietary AA and ω-6 fatty acids increased pro-inflammatory levels in clinical studies (de Batlle et al., 2012; Nielsen et al., 2005). In addition, lower dietary ω-6/ω-3 fatty acids suppressed oral tumors by inhibiting tumor cell proliferation in hamsters (Lee et al., 2018), and dihomo-γ-linoleic acid level contributed to obesity risk and insulin resistance in Japanese subjects with type 2 diabetes (Tsurutani et al., 2018).

However, based on large-scale clinical observational studies and meta-analysis, the American Heart Association suggests that consuming 5–10% ω-6 fatty acids reduce the risk of cardiovascular disease by improving lipoprotein profile (Harris et al., 2009). Likewise, recent research investigated LA consumption impacts on cardiovascular disease improvements via reduced lipid profile (Farvid et al., 2014; Froyen et al., 2020). Thus, the effect of ω-6 fatty acid intake on health is controversial. In this paper, we summarize the recent five years of clinical findings of ω-6 fatty acid effects on certain diseases (e.g., inflammation, cancers, cardiovascular disease, and metabolic syndromes).

Materials and methods

Literature search

We conducted our systematic review according to the Cochrane guidelines and performed the literature search using the PICO strategy: participants (P), intervention (I), comparison (C), and outcomes (O). Participants were healthy at baseline, and the intervention was an amount of dietary ω-6 fatty acids. Any relevant comparisons were accepted: placebo, different fatty acids, different dosage, and baseline. Outcomes were disease-related biomarkers or incidence. We limited the publication period to the past five years (Jan. 2014 to Oct. 2019). We performed the literature search on October 6th, 2020, using PubMed/MEDLINE, Web of Science, EMBASE, Cochrane, and National Digital Science Library (NDSL) databases. Search keywords contained various diseases and ω-6 fatty acids as follows; (stroke OR cardiovascular* OR dementia OR “cognitive dysfunction” OR “cognitive impairment” OR dyslipidemias OR hyperlipidemia OR hypertension OR “blood pressure” OR “diabetes mellitus” OR neoplasms OR cancer OR osteoporosis OR obesity OR obes* OR inflammation OR inflam* OR “lifestyle modification” OR “dietary pattern” OR exercise OR death OR mortality) AND (omega-6 OR “omega 6” OR n6 OR n-6 OR ω-6 OR ω6) with MeSH Terms.

Eligibility criteria

We included randomized controlled trials (RCTs) and cohort studies that assessed relationships between dietary ω-6 fatty acids and diverse diseases. Articles using unsuitable study designs and participants (e.g., patients and animals), abstracts, inappropriate intake methods (e.g., injection), and languages other than English were excluded. Literature published before 2014 was excluded. Two independent individuals conducted literature exclusion in three sessions (title, abstract, full text). If the two independent opinions conflicted, we collected a third opinion. Repeated articles were removed using Endnote X9 (Clarivate Analytics, Philadelphia, PA, USA) and Microsoft Excel (Microsoft, Redmond, WA, USA).

Quality assessment

RCT quality was assessed using Risk of Bias 2.0 (Cochrane Collaboration, Oxford, England). Evaluation items were composed of (1) randomization process, (2) deviations from intended interventions, (3) missing outcome data, (4) outcome measurement, and (5) reported results selection. Results were separated as low risk, some concern, and high risk. Cohort studies were evaluated using the Newcastle–Ottawa Quality Assessment Scale. The assessment included selection, comparability, and outcome domains.

We assessed evidence strength using the 2015 United States Department of Agri-culture (USDA)’s Nutrition Evidence Library conclusion statement evaluation criteria (Alexandria, VA, USA). The strength of evidence was evaluated for bias risk, quantity, consistency, and impact. The assessment grade was divided into strong, moderate, and limited.

Data extraction

We summarized the characteristics of included studies in a table incorporating study design. For example, RCT study results included the following information: outcome, number and health status of subjects, age, proportion of males, comparisons, daily dose and duration, and biomarkers. Cohort studies included outcome, number of subjects, age, proportion of males, follow-up period, exposure and assessment methods, divisions and number of cases, odds and hazard ratios, and adjustment variables.

Results and discussion

Literature search and study selection

We selected literature based on eligibility criteria (Fig.  1). We searched 23,181 records (PubMed = 6,124; Web of Science = 10,020; EMBASE = 1,776; Cochrane = 3,729; NDSL = 1,532) and verified 18,457 studies after removing 3,889 duplicates and 835 publications prior to 2014. Another 18,366 were excluded based on title and abstract review. A total of 91 records was evaluated by full-text review, and 70 articles were excluded according to study design incompatibility (n = 36), irrelevant purpose (n = 21), not an article (n = 5), patient specifics (n = 5), non-oral intake (n = 2), and overlap (n = 1). Of the original 23,181 records searched, we selected 21 studies for further quality assessment.

Fig. 1.

Flow diagram of the literature search and study selection

Quality assessment

Two independent reviewers further examined individual articles for bias risk using methodological quality. Twelve RCT studies were evaluated using the Cochrane tool (Fig. 2). As a result, five studies were assessed as low risk, six studies were assessed as some concern, and one study was assessed as high risk. The six studies of some concern did not specify randomization or allocation methods, three of the studies did not stipulate participant or assessor blinding procedures, and one study, which exaggerated specific results, was assessed as high risk for reported results.

Fig. 2.

Risk of bias assessment in randomized controlled trials. Green, yellow, and red dots mean bias risk value of low, some concern, and high, respectively

Nine cohort studies were examined using the Newcastle–Ottawa scale (Table 1). The Newcastle–Ottawa scale is a validated scale used to assess observational studies. Each item is awarded star points, with a maximum 9 points available: selection—4 points, comparability—2 points, and outcome—3 points. Studies awarded a high score in all items were assessed as low risk. Finally, six studies were evaluated as low risk, two studies were moderate, and one study was high. Studies including assessment of exposure due to self-reporting and unrepresentative participants lost points in the selection part. If the study adjusted for both important and additional factors, maximum star points were given in the comparability section. Moderate and high risk studies also accomplished study by outcome self-reporting. All studies classified nine diseases for assessing the strength of evidence. The five diseases of dyslipidemia, inflammation, cancer, vascular function, and insulin resistance were evaluated as strong evidence, indicating the effects were similar to reality. Three other diseases, including body weight, kidney disease, and glaucoma, were assessed as moderate.

Table 1.

Cohort study quality evaluated using the Newcastle–Ottawa scale

Author, year	Selection	Comparability	Outcomes	Risk of bias
De silva, 2014	***	*	**	Low
Kiyabu, 2015	**	**	**	Moderate
Sellem, 2018	***	*	*	High
Luu, 2018	***	**	**	Low
Jakobsen, 2014	***	**	***	Low
Krishnan, 2017	***	**	***	Low
Dow, 2016	***	*	**	Low
Malhotra, 2016	***	**	***	Low
Pérez de Arcelus, 2014	**	**	**	Moderate

Selection items were (1) exposed cohort representativeness, (2) non-exposed cohort selection, (3) exposure ascertainment, and (4) demonstration that the outcome of interest was not present at study onset. The one comparability item was cohort comparability based on design or analysis. Outcome items were (1) outcome assessment, (2) follow-up long enough for outcomes to occur, and (3) adequacy of cohort follow up

ω-6 Fatty acids and inflammation

External materials invading damaged cells stimulate an inflammatory response (Greten and Grivennikov, 2019). When the immune system detects non-self-materials, immune cells produce cytokines at specific locations (Medzhitov, 2010). PGE2, PGI2, LTB2, TXA2, and TXB2 derived from AA are major mediators of inflammatory response (Innes and Calder, 2018). A study revealed that sunflower oil supplements (2.5 g ω-6 fatty acids) increased TNF-α concentration after a physical load test in 75 young male soccer players (Table 2) (Radoman et al., 2015). Similarly, a cohort study indicated that AA (0.07–0.22 g/d) consumption induced ulcerative colitis over a 3.8-year period (Table 3) (de Silva et al., 2014). These studies, however, presented weak evidence because they were conducted with a small number of unrepresentative participants. There also is insufficient evidence tying dietary ω-6 fatty acids with increased inflammation (Innes and Calder, 2018).

Table 2.

ω-6 Fatty acid intake and disease association in randomized controlled trials

Reference	Outcome	Subject No	Subject health	Age	Male (%)	Comparisons	Daily dose/duration	Markers
Radoman, 2015	Inflammation	75	Healthy	18–19	100	Baseline	2.5 g ω-6 2-months	↑ TNF-α
Markworth, 2018a	Inflammation	19	Healthy	18–35	100	Placebo	1.5 g/day AA 4-weeks	(-) IL-1β, CCL-2, and TNF in PBMC and skeletal muscle
Markworth, 2018b	Inflammation	19	Healthy	18–35	100	Placebo	1.5 g/day AA 4-weeks	(-) IL-1β, TNF, and IL-6 in PBMC and skeletal muscle
Dias, 2016	Hypercholesterolemia	25	Healthy	18–65	21	20.9 g SFA +  2.4 g ω-3	20 g ω-6 +  2.4 g ω-3 6-weeks	↓ TC, LDL, and TC/HDL
Dias, 2017a	Hypercholesterolemia	27	Healthy	21–65	26	20.9 g SFA +  2.4 g ω-320.9 g SFA	20 g ω-6 +  2.4 g ω-3 20 g ω-6 10-days	↓ LDL, LDL-P, and LDL-TG
Dias, 2017b	Hypercholesterolemia	26	Healthy	18–65	20	20.9 g SFA +  2.4 g ω-320.9 g SFA	20 g ω-6 +  2.4 g ω-3 6-weeks20 g ω-6 6-weeks	↓ IDL-TG and LDL-TG
Lyte, 2016	Hypertriglyceridemia	20	Healthy	25	60	10% SFA +  2% ω-6 + 0.5% ω-3	10% SFA + 7% ω-6 1-week	↑ TG and (-) FFA
Young, 2017	Hypertriglyceridemia Hypercholesterolemia	71	Healthy	18–40	52	ω-6:ω-3 = 10	2–3 ω-6:1 ω-3 10-weeks	(-) Plasma TG, TC, LDL, and HDL
Vafeiadou, 2015	Hypercholesterolemia Vascular function	195	Moderate CVD risk	21–60	44	17% total energy SFA	10% total energy ω-6 16-weeks	↓ TC, LDL, and TC/HDL (-) VCAM-1 and ICAM-1 (-) IL-6 and TNF-α
Kim, 2017	Vascular function	147	Healthy	30–65	16	Low LA	High LA 8-weeks	↑ ApoB, ox-LDL, and Lp-PLA2
Rathnayake, 2018	Vascular function	32	Menopausal women	48–65	0	52 g SFA	56.2 g ω-6 1 day	↓ sICAM-1
Mitchell, 2018	Muscle adaptation	19	Healthy	25.5	100	Placebo	1.5 g AA 4-weeks	↑ PAX7 and MYOG ↑ 45S, ITS + 28S, ITS + 5.8S, and 5.8S

AA arachidonic acid, IL-1β interleukin-1β, CCCL-2 chemoattractant protein-2, TNF tumor necrosis factor, PBMC peripheral blood mononuclear cell, SFA saturated fatty acid, TC total cholesterol, LDL low-density lipoprotein, HDL high-density lipoprotein, LDL-P LDL particle, TG triglycerides, FFA, free fatty acids, CVD cardiovascular disease, VCAM-1 vascular cell adhesion molecule-1, ICAM-1 intercellular adhesion molecule-1, LA linoleic acid, ApoB apolipoprotein B, ox-LDL oxidized low-density lipoprotein,IL-1β Lp-PLA₂ lipoprotein-associated phospholipase A2, PAX7 paired box 7, MYOG myogenin, ITS internal transcribed spacer

Table 3.

Prospective cohort study characteristics of dietary ω-6 fatty acid intake and disease risk

References	Outcome	Subjects	Age	Male (%)	Followup (years)	Exposure and assessment method	Divisions and No. of case	Odd or hazard ratio	Adjustments
de Silva, 2014	Ulcerative colitis	130	64.9	58	3.8	AA 7 d food diary	T1 (lowest): 8 T2: 10 T3 (highest): 8	T1: 1 T2: 5.25 T3: 6.09	Energy intake, aspirin use, cigarette smoking, social class, dietary intakes of LA, EPA, DHA, and LNA
Kiyabu, 2015	Breast cancer	38,234	45–74	0	14.1	ω-6 PUFA FFQ	Q1 (lowest): 130 Q2: 139 Q3: 138 Q4 (highest): 149	Q1: 1 Q2: 1.10 Q3: 0.89 Q4: 0.96	BMI, age at menarche, age at first birth, parity, menopausal age, menopausal status at baseline, use of exogenous female hormones, leisure-time physical activity, smoking status, alcohol intake, and total energy-adjusted intake of isoflavones
Sellem, 2019	Digestive cancer	44,039	56.9	27	9	ω-6 PUFA 24-h dietary record	Q1 (lowest): 51 Q2: 37 Q3: 39 Q4: 25 Q5 (highest): 38	Q1: 1 Q2: 0.63 Q3: 0.66 Q4: 0.39 Q5: 0.56	Age, gender, BMI, height, physical activity, smoking status, number of 24-h dietary records, alcohol intake, energy intake without alcohol, family history of cancer, educational level, total lipid intake, and fruit and vegetable intake
Luu, 2018	Lung cancer	121,970	57	47	8.8	ω-6/ω-3 PUFA semi-FFQ	Q1 (lowest): 179 Q2: 175 Q3: 150 Q4: 162 Q5 (highest): 163	Q1: 1 Q2: 0.96 Q3: 0.79 Q4: 0.81 Q5: 0.70	Age, smoking status, smoking pack-years, drinking status, physical activity status, total energy, vitamin supplement use, hormone therapy replacement, and menopausal status
Jakobsen, 2015	Body weight	29,152	55	44	5.3	ω-6 PUFA semi-FFQ	T1 (lowest, β) T2: -19.0 g T3 (highest, β): -189.7 g	-	Sex, age, BMI at recruitment, education, smoking status, leisure-time physical activity, alcohol consumption, and intakes of proteins, long-chain n-3 PUFA, and energy
Krishnan, 2017	Diabetes mellitus	5,288	53.5	38	12	ω-6 PUFA semi-FFQ	-	T1 (lowest): 1 T2: 0.88 T3 (highest): 1.01	Age, sex, race, study center, education, leisure-time physical activity, smoking, alcohol consumption, and BMI
Dow, 2016	Diabetes mellitus	71,334	52.9	0	18	LA and AA 208-item dietary questionnaire	LA T1 (lowest): 740 T2: 838 T3 (highest): 1069 AA T1 (lowest): 579 T2: 750 T3 (highest): 1281	LA T1: 1 T2: 0.98 T3: 0.97 AA T1: 1 T2: 1.11 T3: 1.49	Energy intake, alcohol consumption, education, smoking status, hypertension, hypercholesterolemia, family history of diabetes, physical activity, and BMI
Malhotra, 2016	End-stage renal disease	4,304	54.6	45	11	ω-6 PUFA FFQ	Q1 (lowest): 205 Q2: 217 Q3: 235 Q4: 206 Q5 (highest): 211	Q1: 1.00 Q2: 1.02 Q3: 1.03 Q4: 0.85 Q5: 0.81	Age, sex, race, total energy intake, diabetes, percent energy from protein intake, BMI, hypertension, education level, household income, smoking status, and percent energy from saturated fat intake
Pérez de Arcelus, 2014	Glaucoma	17,128	39	41	8.2	ω-3: ω-6 ratio semi-FFQ	Q1 (lowest): 21 Q2: 25 Q3: 25 Q4: 40 Q5 (highest): 45	Q1: 1.00 Q2: 1.38 Q3: 1.35 Q4: 1.85 Q5: 1.91	Age, sex, BMI, smoking, hypertension, diabetes mellitus, physical activity, coffee consumption, alcohol consumption, adherence to the Mediterranean diet, and total energy intake

AA arachidonic acid, LA linoleic acid, EPA eicosapentaenoic acid, DHA docosahexaenoic acid, LNA α-linolenic acid, FFQ food frequency questionnaire, BMI body mass index, PUFA polyunsaturated fatty acid

Three RCT studies in our review investigated the relationship between ω-6 fatty acid consumption and muscle inflammation (Table 2). AA concentrations in the blood and muscle of healthy young men increased with AA supplementation (1.5 g/d for 4 wks) compared to the placebo. However, inflammatory cytokine mRNA expression, including interleukin-1β (IL-1β), chemoattractant protein-2 (CCL-2), and tumor necrosis factor (TNF), in the peripheral blood mononuclear cells and muscle were unchanged (Markworth et al., 2018a). In a further study, 1.5 g/d AA supplements transiently accelerated acute cytokines (IL-1β, TNF, and IL-6) in skeletal muscle after resistance exercise, but the changes disappeared within 48 h (Markworth et al., 2018b). A previous RCT study reported that 700 mg AA consumption for 12 wks was ineffective in increasing the number of immune cells or cytokine levels (Thies et al., 2001). A group of elderly Japanese subjects who were provided AA-enriched oil (240 or 720 mg/d) for 4 wks did not show a change in inflammatory markers (C-reactive protein, IL-6, and TNF-α) despite an increasing AA concentration (Kakutani et al., 2011). In summary, there was insufficient evidence to conclude that dietary ω-6 fatty acids have a detrimental effect on inflammatory cytokines in clinical studies.

ω-6 Fatty acids and cancer

ω-6 Fatty acids synthesize eicosanoids via cyclooxygenase (COX) and lipoxygenase (LOX) pathways (Schneider and Pozzi, 2011). Eicosanoids are expressed extensively in the body, and the resulting pro-inflammatory response enhances tumor cell proliferation and growth (Wang and Dubois, 2010). However, recent observational studies and review articles have investigated that ω-6 fatty acids are not associated with cancer risk (e.g., breast, prostate, colorectal, skin) (Kim and Kim, 2020). For these reasons, the effect of ω-6 fatty acids on cancer has conflicted. A systematic review reported the relationship between polyunsaturated fatty acids and cancer incidence, which was explained with total cancer diagnosis and death, and breast and prostate cancer diagnosis in 47 RCTs studies, also reported that cancer outcomes are not changed by ω-6 fatty acids (Hanson et al., 2020). In our review, three large-scale cohort studies investigated the effects of fatty acid consumption on existing cancer (Table 3). One study evaluated the effects of ω-6 fatty acid intake on breast cancer in 38,234 Japanese women (aged 57) (Kiyabu et al., 2015). Higher ω-6 fatty acid intake (14.3 g/d; 6.4%/2000 kcal) appeared to induce breast cancer onset, but the tendency was lost after adjustments. A French cohort study revealed that ω-6 fatty acids have no association with breast cancer; the study even determined that 13.7 g/d (6.2%) ω-6 fatty acids consumption leads to digestive cancer prevention (Sellem et al., 2019). Similarly, a previous systematic review indicated that dietary AA consumption had no association with breast or prostate cancer risk (Sakai et al., 2012). The protective effect on digestive cancer was explained by a previous in vitro study that verified that LA accelerated apoptosis by enhancing lipid peroxides in colorectal cancer cell lines (Lu et al., 2010). In addition, a Chinese cohort study showed that lung cancer was prevented through an 8.8-yr follow-up period in 121,970 non-smoking Chinese subjects who consumed a high ω-6:ω-3 fatty acid ratio (7:1) (Luu et al., 2018). However, the positive effect was primarily reported, therefore, a detailed signaling pathway was not verified yet. Collectively, large-scale cohort studies demonstrate that ω-6 fatty acids prevent digestive and lung cancers.

ω-6 Fatty acids and cardiovascular diseases

Dyslipidemia is a dominant risk factor for cardiovascular diseases because the mechanism leading to inflammation, endothelial dysfunction, and lesion formation is the core cardiovascular disease pathophysiology. The association between ω-6 fatty acid intake and dyslipidemia is confirmed through several RCT studies. Co-administration of ω-3 (2.4 g/d) and ω-6 fatty acids (20 g/d; 9%) for 6 wks showed decreased levels of total cholesterol (TC), low-density lipoprotein (LDL), and the ratio of TC:HDL (high-density lipoprotein) compared to co-administration of ω-3 and SFA (Dias et al., 2016). Another study investigated the lipid profiles related to cardiovascular disease using the ω-3 fatty acid pre-supplementation (2.4 g/d) followed by SFA (20.9 g/d) or ω-6 fatty acid (20 g/d). As a result, LDL and LDL subclass distribution (LDL-particle and LDL-TG), measured by nuclear magnetic resonance, were improved with co-administration of ω-3 and ω-6 fatty acids compared to the co-administration of ω-3 and SFA (Dias et al., 2017a). In addition, the lipoprotein profiles (IDL-TG and LDL-TG) improved with 10 wks of co-administration of ω-3 (2.4 g/d) and ω-6 fatty acid supplements (20 g/d) compared to the co-administration of ω-3 (2.4 g/d) and SFA (20.9 g/d) (Dias et al., 2017b). Similarly, in 2009, Health Canada’s Food Directorate replaced SFA with polyunsaturated fatty acids to lower cholesterol (Health Canada, 2012). However, two RCT studies showed that adjusting the ratio of ω-6:ω-3 had no effect on blood lipoproteins. Postprandial serum TG level increased in ω-6 fatty acid enriched-meal group (7% ω-6 and 10% SFA) compared to ω-3 fatty acid enriched-meal group (0.5% ω-3, 2% ω-6, and 10% SFA) (Lyte et al., 2016). Another study demonstrated that plasma lipid profile (TG, TC, LDL, and HDL) was not influenced by the ω-6:ω-3 ratio (10:1 vs. 3:1) in healthy people (Young et al., 2017). An Italian case–control study showed reduced myocardial infarction risk by ω-6 fatty acids, but the beneficial effect of the ratio adjustment was unclear (Marangoni et al., 2014). It has limitation that those studies indicated co-administration of ω-6 and ω-3 fatty acids; therefore, independent effects of ω-6 fatty acids were not investigated. Nevertheless, above results showed that ω-6 fatty acids have effect on reducing lipid profiles, especially cholesterol, when replacing SFA. Taken together, the overall evidence shows that ω-6 fatty acids improve lipoprotein profile when substituting for SFA.

Several studies have reported inconsistent results for vascular function. One study comparing ω-6 fatty acid-rich diets (ω-6 10.5% energy and SFA 8% energy) with SFA-rich diets (ω-6 3.1% energy and SFA 17.6% energy) in moderate cardiovascular disease risk subjects indicated that ω-6 fatty acid-rich diets are ineffective on endothelial activated vascular cells, intracellular adhesion molecules (VCAM-1 and ICAM-1), and inflammation markers (IL-6 and TNF-α) despite improving serum TC, LDL, and TC/HDL (Vafeiadou et al., 2015). In addition, LA (5.4 g/d) supplements enhanced vascular function risk factors, including apoprotein B, oxidized LDL, and lipoprotein-associated phospholipase A2 activity (Kim et al., 2017). However, a ω-6 fatty acid-rich diet (51.3 g/d; 23.1%) reduced postprandial plasma-soluble intercellular adhesion molecule 1 (sICAM-1), which is a major marker of atherosclerosis in menopausal women compared to the SFA (Rathnayake et al., 2018). Recently, the American Heart Association reported ω-6 fatty acid intake to be harmless with respect to cardiovascular disease mortality, although it increases inflammation and oxidative damage (Sanders, 2019). Large-scale (30 observational studies, 13 countries) reviews concluded that LA intake reduced total cardiovascular mortality and ischemic stroke. They also determined that AA was not associated with cardiovascular disease incidence and mortality (Marklund et al., 2019). Thus, ω-6 fatty acids are helpful in preventing cardiovascular disease by improving lipid profiles.

ω-6 Fatty acids and metabolic syndrome

The effects of ω-6 fatty acid supplements on body weight, insulin resistance, and chronic renal diseases were investigated through four cohort studies. A Danish study examined the ratio of carbohydrates to ω-6 fatty acids and showed that body weight and waist circumference were not significantly changed by high amounts of ω-6 fatty acids (6.9% energy) in low-carbohydrate (38.5% energy) groups (Jakobsen et al., 2015). Second, in a diabetes study, hazard ratios of impaired fasting glucose with impaired glucose tolerance were reduced by ω-6 fatty acids 4–5% energy (Krishnan et al., 2017). Another study, however, reported that LA intake was not associated with diabetes, though intake of more than 0.19 g/d AA increased diabetes risk in the form of obesity (Dow et al., 2019). Finally, in a meta-analysis that examined 20 independent cohort studies, high levels of LA in phospholipids prevented diabetes, while AA consumption had no association with diabetes (Wu et al., 2017). Similarly, type 2 diabetes risk was reduced with high LA intake (7.16% energy) (Zong et al., 2019). According to a cross-sectional study, however, a high intake ω-6:ω-3 ratio (> 11.74) led to increased adiposity, waist circumference, and HOMA-IR (Torres-Castillo et al., 2018). Therefore, we conclude that a moderate intake of total ω-6 fatty acids, especially LA, alleviates diabetes risks, but high-doses of fatty acids can lead to diabetes. Hypertension and diabetes accelerate the risk of chronic kidney disease, which can lead to end-stage renal disease (ESRD). Because evidence shows that ω-6 fatty acids alleviate diabetes, a cohort study investigated the relationship between dietary ω-6 fatty acids and ESRD. End-stage renal disease occurrence was reduced for 11 yr in the highest ω-6 intake (10% energy) group after adjustments (Malhotra et al., 2016). Taken together, the findings indicate that ω-6 fatty acid consumption is helpful in delaying diabetes onset and chronic renal disease.

Others

Based on eicosanoid effects, ω-6 fatty acid impacts were explored on muscle adaptation and glaucoma. Muscle protein synthesis is regulated by prostaglandins and COX pathways (Trappe and Liu, 2013). Thus, dietary AA (1 g/d for 50 d) enhances anaerobic performance and improves muscle in resistance-trained males (Roberts et al., 2007). Our review also indicates that AA supplement (1.5 g/d) altered muscle recovery by increasing myogenic genes (paired box 7 and myogenin) and ribosomal RNA (45S, internal transcribed spacer (ITS) + 28S, ITS_5.8S, and 5.8S) expression in the muscle of resistance-trained males (Mitchell et al., 2018). Therefore, the AA supplement aided muscle recovery, which was mediated by resistance training.

Prostaglandins are well-known for suppressing intraocular pressure, a glaucoma risk factor (Stjernschantz, 2001). A previous cohort study investigated whether a high ω-3:ω-6 ratio increased primary open-angle glaucoma by increasing intraocular pressure (Kang et al., 2004). Similarly, our recent cohort studies reported that ω-6 fatty acids delayed the onset of glaucoma in 17,128 participants for the 8.2 year follow-up period (Pérez de Arcelus et al., 2014). In summary, dietary ω-6 fatty acids reduce glaucoma incidence by decreasing intraocular pressure.

This systematic review used recent clinical studies to examine the relationship between ω-6 fatty acids and various diseases. We used 21 articles, including 12 RCTs and 9 cohorts, that were selected by exploring ω-6 fatty acid synonyms and diseases. We classified selected articles into five categories: (1) inflammation, (2) cancer, (3) cardiovascular disease, (4) metabolic syndromes, and (5) others. Three RCT studies reported that ω-6 fatty acid consumption did not contribute to inflammatory responses, indicating that dietary ω-6 fatty acid is not an inflammation risk factor. In several cohort studies, ω-6 fatty acid was not related to the onset of breast cancer and helped prevent digestive cancer and lung cancer. ω-6 fatty acid also alleviated cardiovascular disease via improving cholesterol related blood lipoprotein profiles. Moderate ω-6 fatty acid intake delayed diabetes occurrence. Moreover, it prevented chronic renal disease, affected muscle recovery, and decreased the chance of developing glaucoma by reducing intraocular pressure. In conclusion, ω-6 fatty acids have beneficial effects on digestive and lung cancer, blood lipoprotein profiles, diabetes, chronic renal disease, muscle function, and glaucoma without a negative inflammation response.

Selection items were (1) exposed cohort representativeness, (2) non-exposed cohort selection, (3) exposure ascertainment, and (4) demonstration that the outcome of interest was not present at study onset. The one comparability item was cohort comparability based on design or analysis. Outcome items were (1) outcome assessment, (2) follow-up long enough for outcomes to occur, (3) and adequacy of cohort follow up.

Declarations

Conflict of interest

The authors declare no competing interests.