I really applaud anyone who is trying to make positive changes to their lifestyle, especially by improving their nutrition but, few things annoy me more than massively disseminated misinformation. The most virulent case of this is linoleic acid, an omega 6 (n-6) fatty acid.
Do one google search of "Omega 6's and Inflammation" and you'll find articles on every internet health site, written by MDs/RDs/etc etc, telling you that Omega 6's, like those found in vegetable oils, are virtually the cause of every modern day ailment, from heart disease, to acne and arthritis. They'll cite you information stating that the n-6:n-3 fatty acid ratio is all out of whack.
There seem to be 2 reasons that people claim linoleic acid is pro-inflammatory. 1, because it is a precursor to inflammatory molecules in the body. 2, because it is subject to oxidation, and these oxidative endproducts (reactive carbonyl compounds, ketones, aldehydes) incite an immune response .
1. Linoleic acid is a precursor to Arachidonic Acid, which is a precursor to pro-inflammatory molecules in the body, namely prostaglandins and thromboxanes. One might think, and many do, that increasing dietary intake of Linoleic acid will lead to more arachidonic acid and furthermore, lead to more pro-inflammatory molecules.
This doesn't quite happen in such a simplistic manner, as the process of producing AA from LA is a highly regulated process in the body. Reduction of linoleic acid by 90%, as well as 6-fold increases, do not alter the tissue content of arachidonic acid in adults consuming a Western Diet (1). The Journal of the American Heart Association reports that there is no evidence that dietary linoleic acid intake affects the anti-flammatory effects of omega 3's, and that actually, a combination of omega 3's and omega 6's have the most anti-inflammatory benefits (2). Lastly, the Journal of the Academy of Nutrition and Dietetic's systematic review of randomized controlled trials finds no evidence that the addition of linoleic acid increases inflammatory markers (3).
2. The topic of lipid oxidation often comes up when talking about the inflammatory effects of omega 6's. There was a recent article in Advances in Nutrition by Lawrence that discussed the molecular thinking behind SFA's rather benign effects, and n-6's susceptibility to oxidation and subsequent inflammatory processes that contribute to heart disease (4). Many will cite the Sydney diet heart study (5), which was a re-analysis of the data by Ramsden, that found an increased risk of death due to an increased death of n-6's. This data has been critiqued because the participants used n-6's in the form of Miracle Margarine, which at the time contained high levels of trans fats (6). One can read the comments on the BMJ article and see that there are many flaws with how the data was interpreted, and, even in an ideal world where it is correct, the evidence of harm is relatively weak.
Ramsden's meta-analysis of n-6 fatty acids (7) has also been cited to support the notion that n-6's may be dangerous. This review seems to have been written in response to Mozzafarien et al's meta-analysis (8) . Their meta analysis was critiqued for some of the trials that were included, which reduced SFA and trans fats at the same (9). Ramsden's meta-analysis has been responded to by Willett and Mozaffarien - see here - their essential conclusion is that replacement of SFA's with mixed PUFAs (n-6's and n-3's) is beneficial in CHD risk reduction, and data included in Ramdsen's analysis is in support of this.
While theories about why n-6's may be detrimental in theory sound plausible (indeed, I haven't seen strong responses to the lipid peroxidation arguments), it's important to not get hung up too much on a potential mechanism and consider real life data (indeed, many who argue oxidation is of concern leave out the mechanisms by which PUFAs may be beneficial (12)). Apart from RCT's, observational data continues to support a beneficial effect of n-6's. A recent paper (10) by Mozaffarien's group looked at this issue of linoleic acid and increased risk of death. They took several thousand individuals older than 65, free of CVD and measured the n-6 plasma phospholipids (linoleic acid as well as its derivatives). After 18 years of follow up, those with higher levels of linoleic acid in their plasma had significantly lower risk of death, attributable to CVD, even after controlling for a number of lifestyle characteristics and other variables. Taking levels at a health baseline allowed for them to avoid the effects that the diseased state might have on n-6 phospholipids. In this same issue of Circulation, a systematic review and meta-analysis (11) of Linoleic acid was published, finding that higher intakes of linoleic acid are associated with dose-dependent reducions in CHD risk.
Many of the general concerns about the recommended linoleic acid intakes and the accompanying scientific responses can be found here. The committee's replies are at the bottom.
As a side note, the n-6:n-3 ratio isn't actually supported by that strong of data - see here and here. And as noted by Willett et al in their commentary to Ramsden, the benefits of replacing some SFAs with mixed n-6/n-3's is seen, with ratios ranging from 7-21.
Hopefully that's made you more confident that eating Polyunsaturated Fats (and some in place of SFAs) is not going to cause too much harm. I'm sure the data will be continually updated, and more data will generated. One thing I'd like to point out is that, even in the trials where n-6 supposedly caused harm, it was at much higher levels that are currently consumed - Data from NHANES (see here) shows that adult Americans are eating about 12-19g of linoleic acid/day, with large sex differences. Assuming that one is eating a 2000 kcal diet (most are not), that is only 5.8-8.5% of total kcals coming from n-6. More realistic values for males would put this percentage even lower. Given that studies like Sydney heart upped n-6 intake to 15% of total kcals before even maybe sorta potentially but probably not seeing slight harm, the current American intakes are significantly lower than these levels of concern.
I find that this issue often tends to not be a debate over whether almonds are healthy, but rather whether you should use soybean oil over butter when you're cooking. As always, and as stated in the dietary guidelines, it's not about excluding foods - it's about planning an adequate diet. Replacing some SFAs (oh the battle we could have about 'good' and 'bad' SFAs....) with mixed n-6/n-3 PUFAs, generally, appears to be cardioprotective. If you ask me, mixing almond butter into my plain nonfat greek yogurt is doing me a lot better than just having full fat greek yogurt.
I do, overall, like the general commentary made by those like Ramsden that we need to not lump all fatty acids into one pile. For food based recommendations, i think it's important to be realistic and not get too hung up on individual fatty acids, but in times when we can genetically modify plants to develop oils of differing fatty acid contents to optimize our own health, I think it's important to determine which fatty acids are most health promoting.
1. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3132704/
2. http://circ.ahajournals.org/content/108/2/155.abstract
3. http://www.ncbi.nlm.nih.gov/pubmed/22889633
4. http://advances.nutrition.org/content/4/3/294.abstract
5. http://www.bmj.com/content/346/bmj.e8707
6. http://www.smc.org.au/2013/02/round-up-dietary-fats-and-heart-disease-bmj-experts-respond/
7. http://www.ncbi.nlm.nih.gov/pubmed/21118617
8. http://www.plosmedicine.org/article/info%3Adoi%2F10.1371%2Fjournal.pmed.1000252
9.http://ebm.bmj.com/content/15/4/108.extract
10.http://circ.ahajournals.org/content/early/2014/08/14/CIRCULATIONAHA.114.011590.abstract
11. http://circ.ahajournals.org/content/early/2014/08/26/CIRCULATIONAHA.114.010236.abstract
12. http://jn.nutrition.org/content/135/9/2075.long
http://www.ilsi.org/NorthAmerica/Pages/ViewItemDetails.aspx?WebId=678CA61D-37F3-4E1D-90DB-E8879920CA55&ListId=F5A43102-7B6F-4531-A209-AFA2B711282D&ItemID=189
Do one google search of "Omega 6's and Inflammation" and you'll find articles on every internet health site, written by MDs/RDs/etc etc, telling you that Omega 6's, like those found in vegetable oils, are virtually the cause of every modern day ailment, from heart disease, to acne and arthritis. They'll cite you information stating that the n-6:n-3 fatty acid ratio is all out of whack.
There seem to be 2 reasons that people claim linoleic acid is pro-inflammatory. 1, because it is a precursor to inflammatory molecules in the body. 2, because it is subject to oxidation, and these oxidative endproducts (reactive carbonyl compounds, ketones, aldehydes) incite an immune response .
1. Linoleic acid is a precursor to Arachidonic Acid, which is a precursor to pro-inflammatory molecules in the body, namely prostaglandins and thromboxanes. One might think, and many do, that increasing dietary intake of Linoleic acid will lead to more arachidonic acid and furthermore, lead to more pro-inflammatory molecules.
This doesn't quite happen in such a simplistic manner, as the process of producing AA from LA is a highly regulated process in the body. Reduction of linoleic acid by 90%, as well as 6-fold increases, do not alter the tissue content of arachidonic acid in adults consuming a Western Diet (1). The Journal of the American Heart Association reports that there is no evidence that dietary linoleic acid intake affects the anti-flammatory effects of omega 3's, and that actually, a combination of omega 3's and omega 6's have the most anti-inflammatory benefits (2). Lastly, the Journal of the Academy of Nutrition and Dietetic's systematic review of randomized controlled trials finds no evidence that the addition of linoleic acid increases inflammatory markers (3).
2. The topic of lipid oxidation often comes up when talking about the inflammatory effects of omega 6's. There was a recent article in Advances in Nutrition by Lawrence that discussed the molecular thinking behind SFA's rather benign effects, and n-6's susceptibility to oxidation and subsequent inflammatory processes that contribute to heart disease (4). Many will cite the Sydney diet heart study (5), which was a re-analysis of the data by Ramsden, that found an increased risk of death due to an increased death of n-6's. This data has been critiqued because the participants used n-6's in the form of Miracle Margarine, which at the time contained high levels of trans fats (6). One can read the comments on the BMJ article and see that there are many flaws with how the data was interpreted, and, even in an ideal world where it is correct, the evidence of harm is relatively weak.
Ramsden's meta-analysis of n-6 fatty acids (7) has also been cited to support the notion that n-6's may be dangerous. This review seems to have been written in response to Mozzafarien et al's meta-analysis (8) . Their meta analysis was critiqued for some of the trials that were included, which reduced SFA and trans fats at the same (9). Ramsden's meta-analysis has been responded to by Willett and Mozaffarien - see here - their essential conclusion is that replacement of SFA's with mixed PUFAs (n-6's and n-3's) is beneficial in CHD risk reduction, and data included in Ramdsen's analysis is in support of this.
While theories about why n-6's may be detrimental in theory sound plausible (indeed, I haven't seen strong responses to the lipid peroxidation arguments), it's important to not get hung up too much on a potential mechanism and consider real life data (indeed, many who argue oxidation is of concern leave out the mechanisms by which PUFAs may be beneficial (12)). Apart from RCT's, observational data continues to support a beneficial effect of n-6's. A recent paper (10) by Mozaffarien's group looked at this issue of linoleic acid and increased risk of death. They took several thousand individuals older than 65, free of CVD and measured the n-6 plasma phospholipids (linoleic acid as well as its derivatives). After 18 years of follow up, those with higher levels of linoleic acid in their plasma had significantly lower risk of death, attributable to CVD, even after controlling for a number of lifestyle characteristics and other variables. Taking levels at a health baseline allowed for them to avoid the effects that the diseased state might have on n-6 phospholipids. In this same issue of Circulation, a systematic review and meta-analysis (11) of Linoleic acid was published, finding that higher intakes of linoleic acid are associated with dose-dependent reducions in CHD risk.
Many of the general concerns about the recommended linoleic acid intakes and the accompanying scientific responses can be found here. The committee's replies are at the bottom.
As a side note, the n-6:n-3 ratio isn't actually supported by that strong of data - see here and here. And as noted by Willett et al in their commentary to Ramsden, the benefits of replacing some SFAs with mixed n-6/n-3's is seen, with ratios ranging from 7-21.
Hopefully that's made you more confident that eating Polyunsaturated Fats (and some in place of SFAs) is not going to cause too much harm. I'm sure the data will be continually updated, and more data will generated. One thing I'd like to point out is that, even in the trials where n-6 supposedly caused harm, it was at much higher levels that are currently consumed - Data from NHANES (see here) shows that adult Americans are eating about 12-19g of linoleic acid/day, with large sex differences. Assuming that one is eating a 2000 kcal diet (most are not), that is only 5.8-8.5% of total kcals coming from n-6. More realistic values for males would put this percentage even lower. Given that studies like Sydney heart upped n-6 intake to 15% of total kcals before even maybe sorta potentially but probably not seeing slight harm, the current American intakes are significantly lower than these levels of concern.
I find that this issue often tends to not be a debate over whether almonds are healthy, but rather whether you should use soybean oil over butter when you're cooking. As always, and as stated in the dietary guidelines, it's not about excluding foods - it's about planning an adequate diet. Replacing some SFAs (oh the battle we could have about 'good' and 'bad' SFAs....) with mixed n-6/n-3 PUFAs, generally, appears to be cardioprotective. If you ask me, mixing almond butter into my plain nonfat greek yogurt is doing me a lot better than just having full fat greek yogurt.
I do, overall, like the general commentary made by those like Ramsden that we need to not lump all fatty acids into one pile. For food based recommendations, i think it's important to be realistic and not get too hung up on individual fatty acids, but in times when we can genetically modify plants to develop oils of differing fatty acid contents to optimize our own health, I think it's important to determine which fatty acids are most health promoting.
1. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3132704/
2. http://circ.ahajournals.org/content/108/2/155.abstract
3. http://www.ncbi.nlm.nih.gov/pubmed/22889633
4. http://advances.nutrition.org/content/4/3/294.abstract
5. http://www.bmj.com/content/346/bmj.e8707
6. http://www.smc.org.au/2013/02/round-up-dietary-fats-and-heart-disease-bmj-experts-respond/
7. http://www.ncbi.nlm.nih.gov/pubmed/21118617
8. http://www.plosmedicine.org/article/info%3Adoi%2F10.1371%2Fjournal.pmed.1000252
9.http://ebm.bmj.com/content/15/4/108.extract
10.http://circ.ahajournals.org/content/early/2014/08/14/CIRCULATIONAHA.114.011590.abstract
11. http://circ.ahajournals.org/content/early/2014/08/26/CIRCULATIONAHA.114.010236.abstract
12. http://jn.nutrition.org/content/135/9/2075.long
http://www.ilsi.org/NorthAmerica/Pages/ViewItemDetails.aspx?WebId=678CA61D-37F3-4E1D-90DB-E8879920CA55&ListId=F5A43102-7B6F-4531-A209-AFA2B711282D&ItemID=189
Comments
Post a Comment