Evidence debunking low sodium diet

[CaptainSSO]

http://www.medpagetoday.com/Cardiology/Hypertension/26258

I haven't head anyone talk about this new paper published by the AMA, but it's pretty important stuff.

The bottom line is that the low sodium hypothesis originated based on the thought that since
acute sodium ingestion leads to transient hypertension after consuming it, eating sodium therefore must lead to chronic hypertension. Not a bad idea in theory, but the problem is that's all it is. But sometime around the mid 20th century it started to get treated as a fact.

There was a study done, I don't feel like browsing for it right now but I will if anyone doubts it, that reducing sodium consumption by HALF lead to a tiny drop in systolic pressure, something like 10 mmHg for Type 1 hypertension (the numbers are not exact, I'm writing this from recall but it's close). And think about it--how feasible is it to cut out half of your sodium consumption? Not very. That's hard to do. Paricularly for such a modest gain. Everything's got sodium in it.

Anyway, there's nothing I love more than seeing evidence refute hypotheses that we seem to love to treat as empirical facts.

Edit: If you're wondering why I sound so angsty, it's cause I believe medical science should be based in science, it's something I feel passionately about. If not, then why even have phsycians? Might as well treat illnesses with folk remedies. I hate the arrogance of treating something we're not sure about as if we ARE sure about it. Reminds me of the big shot physicists at the close of the 19th century saying "That's it, we've learned all there is to know about physics."

---

Members don't see this ad.

 

[Shnurek]

Good stuff

 

[JC Denton]

Would love to hear what you think about the low-fat diet for obesity.

 

[DeadCactus]

Guess this is the thread all the cool people chose, so I'll copy and paste from the duplicate thread:

"A study showing that people better able to excrete their sodium load are likely to have better cardiovascular health just doesn't strike me as that exciting. That's especially true when the study population is a young cohort with a homogeneous genetic make-up mostly stemming from a population with a prevalent history of salt in their diet. But maybe one day we can save a few more people from the horrors of actually looking at a food label..."

Don't get me wrong, I hate the voodoo still taught and practiced as medicine as much as you. I'm just not seeing much to get excited about in this one study.

 

[CaptainSSO]

Guess this is the thread all the cool people chose, so I'll copy and paste from the duplicate thread:

"A study showing that people better able to excrete their sodium load are likely to have better cardiovascular health just doesn't strike me as that exciting. That's especially true when the study population is a young cohort with a homogeneous genetic make-up mostly stemming from a population with a prevalent history of salt in their diet. But maybe one day we can save a few more people from the horrors of actually looking at a food label..."

Don't get me wrong, I hate the voodoo still taught and practiced as medicine as much as you. I'm just not seeing much to get excited about in this one study.

It's not just this one study. (Rather, this one study does not form the foundation of my argument).

http://hyper.ahajournals.org/content/46/1/31.full.pdf

Here is an article similar to the one I alluded to. In this case for 63 year olds a drop by roughly half in urinary sodium excretion led to a drop of 10/1 mmHg...and this was for stage 2 hypertension. As discussed in the article, the lower the BP to begin with, the lower the drop due to Na+ restric.

My point is not that this article about the deleterious effects of Na+ restriction is the end of the discussion, it's that Na+ restric. just doesn't appear to be very effective, let alone pragmatic (IMO. There's no way I could ever reduce my sodium intake by half, and I even salt a lot of my foods).

My point is just that Na+ restriction is universally recommended for hypertension TX and it isn't very effective at all, and most importantly, it may not be harmless, and the article I posted appears to offer at least some initial evidence that indeed, Na+ may be deleterious. In my book, a marginally effective treatment that could potentially have a much bigger risk that vastly overshadows it's maximum potential benefit should be scrapped.

Would love to hear what you think about the low-fat diet for obesity.

Heh, sounds like you already know. I will leave that for another thread, because that is a multi-faceted argument and not as succinct as the sodium restriction argument. I've talked about it on SDN before and it wasn't well-received by most, but PM me if you want to discuss it, it's one of my favorite topics. This summer when I have more time I'll put up a thread where I can post of some of the best evidence, it will just take some time to get everything together.

 

[ronaldo23]

Heh, sounds like you already know. I will leave that for another thread, because that is a multi-faceted argument and not as succinct as the sodium restriction argument. I've talked about it on SDN before and it wasn't well-received by most, but PM me if you want to discuss it, it's one of my favorite topics. This summer when I have more time I'll put up a thread where I can post of some of the best evidence, it will just take some time to get everything together.

it's only a multi-faceted argument if you talk to doctors who generally know nothing about nutrition.In reality, the evidence is pretty much unequivocal that dietary fat has nothing to do with the obesity epidemic

 

[CaptainSSO]

it's only a multi-faceted argument if you talk to doctors who generally know nothing about nutrition.In reality, the evidence is pretty much unequivocal that dietary fat has nothing to do with the obesity epidemic

 

[Raryn]

There are a small number of people who are salt sensitive and highly benefit from a low sodium diet.

There is a much larger number of people for whom it generally won't hurt to go down a little from the high levels of sodium your average American consumes, but shouldn't completely reconfigure their diet unless they really want to.

Fat vs protein vs carbohydrates (after a minimum level of each) doesn't really matter nearly as much as total calories. Given that energy is energy and that it's all pretty fungible in the body, just eat a diet that consists of some amount of all three and limit your total amount of intake to a reasonable level and you'll be fine. No matter what the diet is. Caveman vs Mediterranean vs the purple polka dot diet.

Finally, something not mentioned above but rather a pet peeve of mine: High fructose corn syrup is 55% fructose. Table sugar (sucrose) is 50% fructose. There is physiologically pretty much zero difference. Switching out our HFCS for sucrose will not solve anything at all.

 

[CaptainSSO]

^Did you read either of the two articles?

"It generally won't hurt." You're basing that on what? I want facts, not your cursory opinion on things. The article I posted gives potential evidence to the contrary, show me some evidence. I don't know that it generally won't hurt, but I DO know that it generally won't benefit.

"To go down a little." A reduction of HALF of sodium intake lead to a drop of 10/1 mmHg. That's for stage 2 hypertension, systolic greater than 160 mmHg. The reduction in diastolic BP was ONE mmHg. That figure falls the lower the blood pressure is to begin with. So a reduction of "a little sodium" is essentially worthless at lowering BP. It's going to cause a drop in blood pressure of much less than 10/1 mmHg. It DOES take a major reconfiguration of the diet to cause a SMALL change in BP. So, if a major reconfiguration of the diet causes a tiny change in BP, what does a less than major reconfiguration do?
Essentially nothing: http://hyper.ahajournals.org/content/46/1/31.full.pdf

And finally....no, a calorie is not a calorie. The human body is dependent on hormones, it's not a bomb calorimeter. I don't want to get into this right now cause it's a bigger pill to swallow than the low sodium diet, but calories ARE different.

You're exactly the type of person I'm trying to reach with this thread, and instead of reading anything or posting any sort of evidence, you simply dismiss everything and respond with your preconceived general opinion. If I wanted that I'd just talk to the American Heart Association, they're the experts at ignoring evidence and sticking to the status quo.

Do some critical thinking, man!!

 

[link2swim06]

^Did you read either of the two articles?

"It generally won't hurt." You're basing that on what? I want facts, not your cursory opinion on things. The article I posted gives potential evidence to the contrary, show me some evidence. I don't know that it generally won't hurt, but I DO know that it generally won't benefit.

"To go down a little." A reduction of HALF of sodium intake lead to a drop of 10/1 mmHg. That's for stage 2 hypertension, systolic greater than 160 mmHg. The reduction in diastolic BP was ONE mmHg. That figure falls the lower the blood pressure is to begin with. So a reduction of "a little sodium" is essentially worthless at lowering BP. It's going to cause a drop in blood pressure of much less than 10/1 mmHg. It DOES take a major reconfiguration of the diet to cause a SMALL change in BP. So, if a major reconfiguration of the diet causes a tiny change in BP, what does a less than major reconfiguration do?
Essentially nothing: http://hyper.ahajournals.org/content/46/1/31.full.pdf

And finally....no, a calorie is not a calorie. The human body is dependent on hormones, it's not a bomb calorimeter. I don't want to get into this right now cause it's a bigger pill to swallow than the low sodium diet, but calories ARE different.

You're exactly the type of person I'm trying to reach with this thread, and instead of reading anything or posting any sort of evidence, you simply dismiss everything and respond with your preconceived general opinion. If I wanted that I'd just talk to the American Heart Association, they're the experts at ignoring evidence and sticking to the status quo.

Do some critical thinking, man!!

Ehhh the research pretty clearly shows salt sensitivity is highly correlated to gender, race, and age. It doesnt really seem your linked study sorted the populations apart. I think it is likely any results got diluted by the non-salt sensitive population.


Lets repeat that study BUT I only want this population...African American, older, female and overweight. I would beat you money a salt restricted diet would lower BP a sizable amount in these people.

I think American Heart just tries to keep the message simple. Even if it is only going to help 30% of the population...its easier to deliver a single message than add on a bunch of exceptions/inclusions.

 

[CaptainSSO]

^????

It states right in the article that "a greater number of black subjects, known to be more sensitive to the effects of salt restriction, were included in the systolic/diastolic group." That's the group that had a change of 10/1 mmHg. That WAS the salt sensitive population.

And just for the sake of debate assuming your argument is true, and Na+ restric helps 30% of the population lower BP, what's it doing to the other 70%? We don't know. Potentially bad things. We're recommending an ineffective dietary alteration that can be causing harm. So so far we have 2 studies linking sodium restriction to morbidity/mortality for a drop in BP that's almost nothing.

Low salt diet increass insulin resistance: http://www.ncbi.nlm.nih.gov/m/pubmed/21036373/

 

[link2swim06]

^????

It states right in the article that "a greater number of black subjects, known to be more sensitive to the effects of salt restriction, were included in the systolic/diastolic group." That's the group that had a change of 10/1 mmHg. That WAS the salt sensitive population.

And just for the sake of debate assuming your argument is true, and Na+ restric helps 30% of the population lower BP, what's it doing to the other 70%? We don't know. Potentially bad things. We're recommending an ineffective dietary alteration that can be causing harm. So so far we have 2 studies linking sodium restriction to morbidity/mortality for a drop in BP that's almost nothing.

Low salt diet increass insulin resistance: http://www.ncbi.nlm.nih.gov/m/pubmed/21036373/

I didnt see anything about ethnic groups in http://hyper.ahajournals.org/content/46/1/31.full.pdf

Let me pose this question....do you think people are actually getting too little sodium even AFTER they are advised to go on a low sodium diet. If you are in the US that is nearly impossible...everything is covered in salt. Look at average American intake vs what we need http://www.cdc.gov/Features/dsSodium/

People would have to adopts a pretty radical diet in the US to become chronically deficient on sodium.

 

[jdh71]

http://www.medpagetoday.com/Cardiology/Hypertension/26258

I haven't head anyone talk about this new paper published by the AMA, but it's pretty important stuff.

The bottom line is that the low sodium hypothesis originated based on the thought that since
acute sodium ingestion leads to transient hypertension after consuming it, eating sodium therefore must lead to chronic hypertension. Not a bad idea in theory, but the problem is that's all it is. But sometime around the mid 20th century it started to get treated as a fact.

There was a study done, I don't feel like browsing for it right now but I will if anyone doubts it, that reducing sodium consumption by HALF lead to a tiny drop in systolic pressure, something like 10 mmHg for Type 1 hypertension (the numbers are not exact, I'm writing this from recall but it's close). And think about it--how feasible is it to cut out half of your sodium consumption? Not very. That's hard to do. Paricularly for such a modest gain. Everything's got sodium in it.

Anyway, there's nothing I love more than seeing evidence refute hypotheses that we seem to love to treat as empirical facts.

Edit: If you're wondering why I sound so angsty, it's cause I believe medical science should be based in science, it's something I feel passionately about. If not, then why even have phsycians? Might as well treat illnesses with folk remedies. I hate the arrogance of treating something we're not sure about as if we ARE sure about it. Reminds me of the big shot physicists at the close of the 19th century saying "That's it, we've learned all there is to know about physics."

maybe some day we can feed all this science into a computer and we won't even need doctors anymore

 

[facetguy]

Fat vs protein vs carbohydrates (after a minimum level of each) doesn't really matter nearly as much as total calories. Given that energy is energy and that it's all pretty fungible in the body, just eat a diet that consists of some amount of all three and limit your total amount of intake to a reasonable level and you'll be fine. No matter what the diet is. Caveman vs Mediterranean vs the purple polka dot diet.

Eventually this notion will go away...but it will take a good while. Getting over the low-fat doctrine will take even longer.

 

[facetguy]

Heh, sounds like you already know. I will leave that for another thread, because that is a multi-faceted argument and not as succinct as the sodium restriction argument. I've talked about it on SDN before and it wasn't well-received by most, but PM me if you want to discuss it, it's one of my favorite topics. This summer when I have more time I'll put up a thread where I can post of some of the best evidence, it will just take some time to get everything together.

Let's not save all that fun for PMs. Air it out.

I just happened to see this today; it's somewhat related:
http://www.npr.org/blogs/money/2012/02/03/146356117/who-killed-lard
How Procter & Gamble convinced everyone that their hydrogenated cottonseed oil Crisco was good for us, while lard was bad for us. Sad thing is, many people (physicians included) still believe this.

 

[jdh71]

Eventually this notion will go away...but it will take a good while. Getting over the low-fat doctrine will take even longer.

Word.

The lack of nutrition taught in medical school is horrifying. I think I got like an hour?

It's very temping to apply a simple concept like basic thermodynamics to the body's energy. Unfortunately this is a horribly ignorant way to approach the subject in a system evolved preferential use of macronutrient for energy.

 

[facetguy]

http://www.youtube.com/playlist?list=PL0A01EBCE2711A5A5
This is a good video series about low carb, lipids, etc. It's an interview of (an apparently well-respected) lipidologist Thomas Dayspring, MD and Gary Taubes. It runs about 1.5 hours.

I would recommend starting with Video 6, which shows Dr. Dayspring running through a lipids discussion, then hit Video 7 for a case study that continues the lipids discussion as well as use of certain meds. Then go back to Video 1 at the beginning and watch the other videos.

If you aren't familiar with this issue, these videos will be a good use of time to get up to speed.

 

[achamess]

http://www.youtube.com/playlist?list=PL0A01EBCE2711A5A5
This is a good video series about low carb, lipids, etc. It's an interview of (an apparently well-respected) lipidologist Thomas Dayspring, MD and Gary Taubes. It runs about 1.5 hours.

I would recommend starting with Video 6, which shows Dr. Dayspring running through a lipids discussion, then hit Video 7 for a case study that continues the lipids discussion as well as use of certain meds. Then go back to Video 1 at the beginning and watch the other videos.

If you aren't familiar with this issue, these videos will be a good use of time to get up to speed.

Excellent video. Taubes is the single best thing to happen to nutritional science in the last few years. I've been in contact with him trying to get him to come to my school. I'm so frustrated when I'm in clinic and obese patients (like 75% of our patient population) come in and the doctor or PA goes "you should make some lifestyle changes. Eat less and exercise more. Eat low fat foods." Ughhhh. How's that worked so far? Last I checked, the country got fatter as dietary fat consumption has gone down, per the recommendations of the 'experts'. I wish I could speak up, but I'm just the lowly med student. What I usually do is write down Why We Get Fat by Gary Taubes on a piece of paper and tell the patient to read it as they leave the room. I don't know how many have, but I'm hoping it's a couple.

Good discussion guys. Keep it up. It's amazing how much nutritional 'science' is obscured by hypotheses that certain people want to be true but the evidence doesn't support. It's 2012. We should know by now what optimal eating should consist of.

 

[Charles_Carmichael]

Is there a good review article that covers this nutrition stuff you guys are talking about? I've read some of this on blogs, but am generally cautious about absolutely believing what's written in a blog. I tried searching on PubMed but couldn't find a good, general review article on this stuff. Maybe you guys who are better-versed on nutrition research know of one? Thanks!

 

[DeadCactus]

My point is just that Na+ restriction is universally recommended for hypertension TX and it isn't very effective at all, and most importantly, it may not be harmless, and the article I posted appears to offer at least some initial evidence that indeed, Na+ may be deleterious. In my book, a marginally effective treatment that could potentially have a much bigger risk that vastly overshadows it's maximum potential benefit should be scrapped.

Fair enough. I can agree that medicine has a terrible habit of universally recommending treatments, screenings, and other interventions that should be far more targeted. I think it's a stretch to say there may be a negative effects with salt restriction based on the inverse relationship between excretion and cardiac events when salt excretion is affected so heavily by factors beyond salt intake but sure maybe there is.



Either way, the lipids/nutrition debate is far more pressing in my opinion. The perpetuation of the calories in vs calories out mantra while ignoring the impact caloric sources have on metabolic hormones and metabolism is holding back the fight against our major public health problem. I'd rather see this thread fully derail and start posting some strong articles on the subject...

 

[Raryn]

^Did you read either of the two articles?

"It generally won't hurt." You're basing that on what? I want facts, not your cursory opinion on things. The article I posted gives potential evidence to the contrary, show me some evidence. I don't know that it generally won't hurt, but I DO know that it generally won't benefit.

"To go down a little." A reduction of HALF of sodium intake lead to a drop of 10/1 mmHg. That's for stage 2 hypertension, systolic greater than 160 mmHg. The reduction in diastolic BP was ONE mmHg. That figure falls the lower the blood pressure is to begin with. So a reduction of "a little sodium" is essentially worthless at lowering BP. It's going to cause a drop in blood pressure of much less than 10/1 mmHg. It DOES take a major reconfiguration of the diet to cause a SMALL change in BP. So, if a major reconfiguration of the diet causes a tiny change in BP, what does a less than major reconfiguration do?
Essentially nothing: http://hyper.ahajournals.org/content/46/1/31.full.pdf

And finally....no, a calorie is not a calorie. The human body is dependent on hormones, it's not a bomb calorimeter. I don't want to get into this right now cause it's a bigger pill to swallow than the low sodium diet, but calories ARE different.

You're exactly the type of person I'm trying to reach with this thread, and instead of reading anything or posting any sort of evidence, you simply dismiss everything and respond with your preconceived general opinion. If I wanted that I'd just talk to the American Heart Association, they're the experts at ignoring evidence and sticking to the status quo.

Do some critical thinking, man!!

Word.

The lack of nutrition taught in medical school is horrifying. I think I got like an hour?

It's very temping to apply a simple concept like basic thermodynamics to the body's energy. Unfortunately this is a horribly ignorant way to approach the subject in a system evolved preferential use of macronutrient for energy.

Except the evidence has shown that pretty much any diet is interchangable when it comes to actual results. High protein, low protein, low fat, high carb, no carb, whatever. Keep track of your calories and your diet will be just as effective as any of the above. What I see here is a lot of rhetoric with not much meat.

And my statement "going down a little won't hurt" is based on a variety of other studies that I've seen over the years that estimated the effect of our increased sodium intake on the incidence of CAD/stroke. It certainly wouldn't hurt your blood pressure to eat a reasonable quantity and it certainly can't be good for your kidneys if you suddenly decide say, 6g/day of Na is perfectly fine. Also, you can't deny that many people specifically have salt sensitive hypertension and see a pretty damn good response when they cut down their Na consumption.

When it comes to advising patients regarding lifestyle changes, I tell them the best way to do it is in baby steps. Switch over to non-calorie containing beverages. Try to keep track of how much you eat (especially snacks). Snack less. Cut down on the most calorie dense foods. Maybe replace them with something like some fish or some reasonable vegetables. Patients are far more likely to respond to advice like that than advice to suddenly switch from everything they've been eating their whole lives to a whole new diet of your own concoction. (Oh, and I tell them to ignore dietary cholesterol, since that has !@#$-all to do with your actual bodies cholesterol, >99% of which is manufactured in your liver anyway.)

 

[DeadCactus]

You're going to complain about rhetoric when you argue based on calorie in vs calorie out? I don't see you citing any more studies than anyone else.

And then you admit that your entire strategy is basically to have them cut out simple carbs...

 

[CaptainSSO]

maybe some day we can feed all this science into a computer and we won't even need doctors anymore

?

There are studies out there backing up what I'm saying, I've posted
3 already. I don't understand your point here, jdh.

 

[Raryn]

You're going to complain about rhetoric when you argue based on calorie in vs calorie out? I don't see you citing any more studies than anyone else.

And then you admit that your entire strategy is basically to have them cut out simple carbs...

Only reason I tell them to cut down simple carbs is because that's where a lot of people's excess calories come from. If you're drinking 1.5L of regular soda every day, it won't really change your energy balance much to eat more fish and less prime rib, will it? Furthermore, one of the actual problems with foods that have simple vs complex carb intake is low fiber, and fiber has been shown over and over again to have tons of benefits for your health. Calorie density is calorie density, and I don't care where the reduction in extra density is coming from, whether it's lean protein vs fatty protein, simple vs complex carbs, or whatever you want.

Here's a trial that compared four different diets and their effects on weight loss over two years: http://www.ncbi.nlm.nih.gov/pubmed?term=19246357 There's a ton of other ones if you want to look, but pretty much every single trial of any mid-long term duration in humans has shown that the major determinant in success re: weight loss is dietary adherence, independent of which diet you pick. The best advice you can give your patients is to ask them their preferences and help advise them in that regard, not try to pigeonhole everyone into your diet of choice. (Short term studies 6 months or less will usually have the high-protein low-carb diet win. But that's not sustainable)

The reason I haven't been citing papers re:sodium is because I really don't have the time to parse through the volumes of literature on the subject for your benefit. There are papers that have shown lowering sodium consumption reduces death rates over time. Those are especially hard studies to do well though.

There are other papers that show the modest reduction in BP for healthy people restricting sodium as stated above are more pronounced in people with hypertension. Even "just" a 5mmHg systolic decrease over years can substantially lower your risk of heart disease, and that's the decrease that has been found over and over again in studies of hypertensive patients.

Furthermore, said modest reduction probably gets even more pronounced over time, as there have been indications that a sodium restricted diet over years decreases how much your BP goes up as you age.

Oh, and sodium restriction makes pretty much every single class of antihypertensive medication we have work better (with the sole exception of calcium channel blockers).

The best summary is the one I tell patients and that I've heard my preceptors tell patients: Will eating less salt be a magic bullet to cure everything? No. Has it been shown to help hypertension? Yes, though more for some people than others. Is paying attention to what you eat a good thing? Yes. Will eating less salt cause any adverse effects? Not unless you go really, really crazy with it, to levels less than 400-500mg a day. Which no one advocates.

 

[CaptainSSO]

"They are especially hard studies to do well."

Whenever someone has to resort to using this, generally the problem is not with the study, it's with the hypothesis.

"Reduction of 5 mmHg can substantially lower your risk..."

Where did you get that?! Show me ANYTHING that backs that up.

I've just shown 2 papers where eating less salt causes adverse affects.

http://www.ncbi.nlm.nih.gov/m/pubmed/21036373/

Everything you've told me, someone in 1950 could have told me. It's 2012. Look at that paper--low sodium diets associated with insulin resistance. Well that's pretty obviously bad, no matter what you believe.

 

[DeadCactus]

So you take a group of obese people weighing 200 lbs give them a 2 year diet and exercise regime complete with regular meetings and, if you're lucky, the 80% that stick around will lose ~10 lbs.

Meanwhile the range of the macronutrient goals accross the different diets is 35-65% fats, 15-25% protein, and 20-40% carbs but the actual intake at 6 months and 2 years is 19.1-42.1% fats, 14.2-27% protein, 34.6-68.6% carbs.

So yea, great study comparing low fat, high carb diets to themselves in a study design that focuses on calories in vs calories out as opposed to the impact of regulatory hormones. The only lesson to take away from that study is that we suck at weight loss management. Ten pounds in two years with a full medical study supporting you? No wonder everyone's fat...

 

[CaptainSSO]

So you take a group of obese people weighing 200 lbs give them a 2 year diet and exercise regime complete with regular meetings and, if you're lucky, the 80% that stick around will lose ~10 lbs.

Meanwhile the range of the macronutrient goals accross the different diets is 35-65% fats, 15-25% protein, and 20-40% carbs but the actual intake at 6 months and 2 years is 19.1-42.1% fats, 14.2-27% protein, 34.6-68.6% carbs.

So yea, great study comparing low fat, high carb diets to themselves in a study design that focuses on calories in vs calories out as opposed to the impact of regulatory hormones. The only lesson to take away from that study is that we suck at weight loss management. Ten pounds in two years with a full medical study supporting you? No wonder everyone's fat...

Very good observations.

"Great study comparing low-fat, high-carb diets to themselves."

Nailed it.

 

[MilkmanAl]

Wait...so if we're saying that calorie type and amount and regular exercise have minimal effect on weight loss, what does affect it? I see hormonal influences mentioned, but the attitudes here seem almost defeatist and/or apologist. That is, it kind of sounds like you guys are saying that people who are fat were going to be fat regardless of lifestyle and diet, while skinny people were going to be the way they are, too. I know this is the hard-line evidence-based-only thread, but do you really need studies to tell you that energy out>energy in is, at the very least, a decent place to start where getting healthy is concerned? Like jdh said, nutrition education in med school is painfully lacking, but everything I've learned is more in line with what Raryn is saying. Maybe it's outdated and unfounded 1950's dogma, but maybe it's been unquestioned for so long because it's obviously correct.

 

[CaptainSSO]

^He was saying, that in fact that study was NOT representative of low carb diets. The study was essentially comparing varying low fat, high carb diets to themselves.

Everythig I learned my whole life was in line with what Raryn was saying as well, until someone educated me on these issues. That's why it's so hard to accept--it runs counter to what we've been taught our whole lives...for a variety of reasons, many political. Anyway I'll make my thread this summer with a LOT of evidence when I have some free time. Gary Taubes's work is one of the best places to start, IMO.

 

[DeadCactus]

Wait...so if we're saying that calorie type and amount and regular exercise have minimal effect on weight loss, what does affect it? I see hormonal influences mentioned, but the attitudes here seem almost defeatist and/or apologist. That is, it kind of sounds like you guys are saying that people who are fat were going to be fat regardless of lifestyle and diet, while skinny people were going to be the way they are, too. I know this is the hard-line evidence-based-only thread, but do you really need studies to tell you that energy out>energy in is, at the very least, a decent place to start where getting healthy is concerned? Like jdh said, nutrition education in med school is painfully lacking, but everything I've learned is more in line with what Raryn is saying. Maybe it's outdated and unfounded 1950's dogma, but maybe it's been unquestioned for so long because it's obviously correct.

No, I think energy out > energy in is a terrible starting place. It's like approaching learning from the time spent sitting in a lecture. The GI system is a tube, not a hole. The intake, absorption, metabolism, and storage of energy are all heavily regulated by hormonal and neural signals. Those signals are modified by multiple inputs including the type of energy being taken in, emotion, genetics, epigenetics, etc.

My problem with the calories in vs calories out mantra is that it focuses on what, to me, is most likely a symptom rather than the problem. We're not garbage disposals. We're pretty damn well self-regulated machines, so why is the focus not on figuring out where our self-regulation is failing?

I can buy that we're a little chunkier because we're more sedentary and food is more readily available. But to dive so far off the metabolic track as to have an obesity epidemic and ubiquitous diabetes? That takes something more than an imbalance between available food and mandated exercise, that takes a significant detour from our normal regulatory systems...

 

[danzman]

No, I think energy out > energy in is a terrible starting place. It's like approaching learning from the time spent sitting in a lecture. The GI system is a tube, not a hole. The intake, absorption, metabolism, and storage of energy are all heavily regulated by hormonal and neural signals. Those signals are modified by multiple inputs including the type of energy being taken in, emotion, genetics, epigenetics, etc.

My problem with the calories in vs calories out mantra is that it focuses on what, to me, is most likely a symptom rather than the problem. We're not garbage disposals. We're pretty damn well self-regulated machines, so why is the focus not on figuring out where our self-regulation is failing?

I can buy that we're a little chunkier because we're more sedentary and food is more readily available. But to dive so far off the metabolic track as to have an obesity epidemic and ubiquitous diabetes? That takes something more than an imbalance between available food and mandated exercise, that takes a significant detour from our normal regulatory systems...

I disagree. You can try to make this as complicated as you want with hormonal variability and epigenetics, and there certainly could be some truth to some of it. But there is a very simple equation at hand here, and if you put someone on an island and feed them fish and fruit for a month they WILL lose weight. It seems that this is a simple case of being penny wise but pound foolish (pardon the pun). Look at the TV show The Biggest Loser. They take huge people, put them on diets, and make them work out. And look what happens, the pounds literally melt off of them!

The average American eats like crap. They eat too much and they eat too many calories. You want people to lose weight get them to cut back on sugar and work out more. We autoregulate just fine, too much glucose --->more fat, Too little glucose--->less fat.

 

[Raryn]

I honestly don't have the time to dedicate to this topic, but I'll give some good starting points.

"They are especially hard studies to do well."

Whenever someone has to resort to using this, generally the problem is not with the study, it's with the hypothesis.

Mortality based studies for noncommunicable diseases are all notoriously hard to do. Hence the debates on mammography. Hence why the first big study to show true statistical benefits of colonoscopy came out last week. This is for a good many reasons including the need for large sample sizes (on the order of thousands ), the need for quite long-term followup, and (if you wanted a study with a high level of evidence) the difficulties and ethics involved in randomizing subjects. Even so, such studies have been done, and while some show no effect, meta-analyses that take advantage of the statistical power of larger sample sizes have shown substantially decreased mortality when you decrease salt consumption. www.bmj.com/content/339/bmj.b4567 is one looking at population studies. http://www.sciencedirect.com/science/article/pii/S0140673611611744 is an analysis of the available randomized controlled trials that addresses many of the difficulties in putting these studies together.

"Reduction of 5 mmHg can substantially lower your risk..."

Where did you get that?! Show me ANYTHING that backs that up.

Looking *only* at death from MI, stroke, or other vascular causes, ignoring all the various other cardiac and renal benefits of lowering blood pressure, the first paper that comes to mind is this meta-analysis from the lancet: http://www.sciencedirect.com/science/article/pii/S0140673602119118 It broke down death rates into blocks of 10 systolic or 5 diastolic mmHg, but it very clearly showed *any* decrease in BP has a significant decrease in mortality. Figures 2 and 4 are the easiest to see this in.

Quote from the conclusions of the paper: "Not only do the present analyses confirm that there is a continuous relationship with risk throughout the normal range of usual blood pressure (down at least as far as 115/75 mm Hg), but they also demonstrate that within this range the usual blood pressure is even more strongly related to vascular mortality than had previously been supposed."

So yes, even 5mmHg matters, assuming you are at at all at risk. And you can find a dozen or more studies that show a low sodium diet in someone who is hypertensive can lower their BP by 5mmHg.

I've just shown 2 papers where eating less salt causes adverse affects.

http://www.ncbi.nlm.nih.gov/m/pubmed/21036373/

Your study of 7 days of dietary modification in 152 healthy people is interesting, and certainly something that should be followed up. I look forward to hearing more about it as they investigate further into possible mechanisms, long term effects, and real health implications related to this.

Everything you've told me, someone in 1950 could have told me. It's 2012. Look at that paper--low sodium diets associated with insulin resistance. Well that's pretty obviously bad, no matter what you believe.

I'm sorry that much of the evidence related to salt is consistent with much of the early evidence from the 1950s. We learn more and change our recommendations as new studies with high levels of evidence come in, but we all stand on the shoulders of giants.

 

[Raryn]

So you take a group of obese people weighing 200 lbs give them a 2 year diet and exercise regime complete with regular meetings and, if you're lucky, the 80% that stick around will lose ~10 lbs.

Meanwhile the range of the macronutrient goals accross the different diets is 35-65% fats, 15-25% protein, and 20-40% carbs but the actual intake at 6 months and 2 years is 19.1-42.1% fats, 14.2-27% protein, 34.6-68.6% carbs.

So yea, great study comparing low fat, high carb diets to themselves in a study design that focuses on calories in vs calories out as opposed to the impact of regulatory hormones. The only lesson to take away from that study is that we suck at weight loss management. Ten pounds in two years with a full medical study supporting you? No wonder everyone's fat...

http://www.ncbi.nlm.nih.gov/pubmed?term=16476868 is another meta-analysis of five studies comparing low-carb to low-fat diets in real world implementations. No measurable difference at 1 year, and each have their own pluses and minuses for metabolic implications.

Looking for any study done on humans with a mid-long term followup that compares different diets and actually finds a difference, I found http://www.ncbi.nlm.nih.gov/pubmed?term=18635428 which does support low carb, though the study has limitations they acknowledge in the discussion.

When a humongous preponderance of evidence says that the most important factor in weight loss is dietary compliance and there's one or two that state maybe the low-carb Atkins is a better option overall, it's still pretty easy to argue that the best thing for them is choosing a diet they would find easier to comply with.

 

[ushush]

^????

It states right in the article that "a greater number of black subjects, known to be more sensitive to the effects of salt restriction, were included in the systolic/diastolic group." That's the group that had a change of 10/1 mmHg. That WAS the salt sensitive population.

And just for the sake of debate assuming your argument is true, and Na+ restric helps 30% of the population lower BP, what's it doing to the other 70%? We don't know. Potentially bad things. We're recommending an ineffective dietary alteration that can be causing harm. So so far we have 2 studies linking sodium restriction to morbidity/mortality for a drop in BP that's almost nothing.

Low salt diet increass insulin resistance: http://www.ncbi.nlm.nih.gov/m/pubmed/21036373/

I did not read your article. But just so you know we are not taught in school that low salt diet is some magic pill. If you do not have hypertension and a family history there is obviously no need to be on any diet. And if you do you are sick whether you are on a diet or not. You can see in Guyton Physiology and similar books that since the 1970's or so there were experiments showing that in some(significant % of) people high salt diet caused much greater changes than in others. Also it is known that hypertension is related to kidney damage. A large % of people with hypertension do not have a reduced level of renin(as they should), which impairs their ability to excrete excess salt/water. Thats why some people also respond to diuretics differently based on their levels of salt intake. Even if you do not have hypertension, your kidney must work harder when you are obese or volume overloaded(ingesting a lot of salt) just like when you have diabetes. Increased GFR is dangerous.
Finally, you mentioned a 10point decrease in BP. I guess I expected greater changes in salt-sensitive people. I remember about some study where people ate only fruits and vegetables for a week. They had very dramatic changes. Maybe in this article that you posted they lived off salt-free ice cream? Isn't the goal of any diet to lose weight? Weight is certainly an important factor in GFR and hypertension.

Btw, no one in my family has hypertension. So this disease is foreign to me. But I feel if I tried to increase my salt intake it would involve eating at McDonalds. And salt is not the only ingredient that makes me sorry for people at mcdonalds. I am sitting at mcdonalds right now and some of the people here just make a lot of bad decisions in their lives that have very little to do with salt or hypertension.

 

[grasshopperking]

I'm glad to see people talking about assumption that dietary fat causes obesity. excess potatoes and sugar cause obesity. Taubes should be standard reading.

 

[CaptainSSO]

Raryn,

Thanks for posting those.

My initial problem with your first article, the meta-analysis, is that in my quick read-through of the article, it suggests 10 of the cohorts had a positive correlation b/w Na+ and morbidity/mortality whereas another 4 had the inverse correlation, I.e. less Na+, more adverse health events. I would need to read into the article more thoroughly but this fact alone suggests to me that something is not quite right. This is not an entirely apt analogy but just imagine for a minute that this same study was done with cigarettes rather than salt--we certainly couldn't implicate cigarettes in this case, the results are conflicting. Anytime there are conflicting results more investigation is needed.

As to your Lancet article, I can't access it, but it looks to me as if their conclusion was that systolic drops of at least 20 mmHg were needed for significant drops (x2) in health risks? I'll have to see if we have access to the Lancet at school in order to look at the Figures 2 and 3 you mentioned.

 

[Asp]

While I think Gary Taubes has done great work, I would say he's done a better job debunking the saturated fat and cholesterol hypothesis than constructing a correct alternative. Fat has been vilified unjustly in the past. Let us not be so quick to now crucify carbs. The explanation is more complicated than abuse of a single macronutrient.

 

[nitemagi]

I always liked this article as the counterpoint to Only looking at sodium -- namely that it's the sodiumotassium ratio in our diet that is a problem.

http://www.ncbi.nlm.nih.gov/pubmed/17494929

 

[thedoctor8706]

I don't eat a low sodium diet because of hypertension... Too much sodium screws with ADHD

 

[Rothbard]

nm

 

[facetguy]

Wait until you folks start re-thinking statins and cholesterol.

Waaaaaayyyy too much money on the line to let that happen.

 

[mTOR]

Wait until you folks start re-thinking statins and cholesterol.

Seriously.

[YOUTUBE]6anV4z5Oi-k[/YOUTUBE]

But simply looking at it from an evidence based standpoint, even just the NNT for statins makes their pervasive usage absolutely silly. But when there's money to be made, who cares about evidence?

Anyways I support this thread. Good to see this stuff discussed.

maybe some day we can feed all this science into a computer and we won't even need doctors anymore

?

There are studies out there backing up what I'm saying, I've posted
3 already. I don't understand your point here, jdh.

lmao. pretty sure it was a side jab my direction.

 

[ToldYouSo]

Wow. A lot of the ideas I've had about diet and nutrition aren't sitting so well anymore (all calories are equal). Feel like there are so many conflicting opinions and studies, gonna watch the Taubes playlist and see what he has to say. Hope the discussion continues

 

[facetguy]

Dietary advice to physicians from Medscape. If it's a low fat diet, it's recommended. If it's not low fat, to the bottom of the list it goes. No surprises.

http://www.youtube.com/watch?v=Uk3V-VLOAc0


But, perhaps Weight Watchers is low carb afterall?
http://waroninsulin.com/nutrition/why-weight-watchers-is-actually-a-low-carb-diet

 

[D elegans]

.

 

[Raryn]

Raryn,

Thanks for posting those.

My initial problem with your first article, the meta-analysis, is that in my quick read-through of the article, it suggests 10 of the cohorts had a positive correlation b/w Na+ and morbidity/mortality whereas another 4 had the inverse correlation, I.e. less Na+, more adverse health events. I would need to read into the article more thoroughly but this fact alone suggests to me that something is not quite right. This is not an entirely apt analogy but just imagine for a minute that this same study was done with cigarettes rather than salt--we certainly couldn't implicate cigarettes in this case, the results are conflicting. Anytime there are conflicting results more investigation is needed.

As to your Lancet article, I can't access it, but it looks to me as if their conclusion was that systolic drops of at least 20 mmHg were needed for significant drops (x2) in health risks? I'll have to see if we have access to the Lancet at school in order to look at the Figures 2 and 3 you mentioned.

Once again, mortality studies of anything are notoriously hard to do well, but the overall conclusions in those meta-analyses regarding Na I linked are fairly sound. Feel free to look at the methodology, they were pretty thorough.

As for the Lancet article re: blood pressure, it was pretty much saying that the risks went up pretty linearly with BP. So changing from 115 to 135 systolic would double your risk, going to 155 would double it again, and going to 175 would give you a 6x risk. Etc. The trend held when you broke it down into smaller quanta, so how many mmHg you considered significant depends on how much value you place on the increased risks. If you think 1.5x risk is significant, then a value of 10mmHg is something to worry about. Etc.

The JNC 7 guidelines have it mostly right from current evidence, though many people can argue that the huge secondary benefits of ACE I's make them a better first line drug than the thiazide-type diuretics. Honestly, the only reason thiazides were first line in those guidelines was because they were generic in 2003 when the guidelines came out. Since the ACEs are becoming generic, there's really no reason not to use them preferentially. Evidence also conflicts regarding what the appropriate blood pressure goals are for diabetics, as well as the best way to treat people in the "prehypertensive" range. JNC 8 will come out at some point and will hopefully address these and other concerns, but you can trust me that Na restricted diets will still be a recommendation.

Wait until you folks start re-thinking statins and cholesterol.

Lol. Because a multitude of randomized controlled trials consisting of tens of thousands of people on statins that show decreases in mortality (from heart attack, stroke, as well as all cause mortality) doesn't mean anything right? Works better in some sub-populations than others, but personally, I think a decreased risk of death is a good thing. (Note: this is for reasonably dosed statins. High dose statins as being better for everybody was a theory that did not hold up, and medical practice has changed. There are some early studies that had increased risk of mortality, but newer statins at studied doses lower your risk of death substantially)

Even with *brand name* statins, cost effectiveness for *primary prevention* (not even secondary prevention) is <$25k per quality-adjusted life year gained, where even the UK's NHS (via NICE) uses ~$50k as the point where they call something cost effective.

Now, the effectiveness of all other non-statin cholesterol lowering medications on actually decreasing mortality is rather suspect, and many experts recommend that even if a statin alone isn't enough, using other medications to change the numbers doesn't seem to decrease mortality (or may even increase it). This is something still up for debate (as are high dose statins in certain higher risk populations), and is something that won't be clear until we have more studies about the non-statin drugs. (Funnily enough, the lipidologist in the video series linked above is apparently in the "treat aggressively with combinations of drugs" camp. At least in one of the later videos. He's welcome to that opinion, but it isn't the most evidence based one in the world)

Jeebus guys. I love how you are so quick to throw out real medical evidence justified by the highest levels of literature available because of some videos of a journalist/blogger you saw on youtube. Do some people exaggerate the benefits of things like a low Na diet or statin use? Sure. But to say that they aren't helpful? I hope that most of you are first or second years that haven't had any family or outpatient IM months yet, because otherwise, your professors have been doing you a real disservice.

 

[danzman]

Once again, mortality studies of anything are notoriously hard to do well, but the overall conclusions in those meta-analyses regarding Na I linked are fairly sound. Feel free to look at the methodology, they were pretty thorough.

As for the Lancet article re: blood pressure, it was pretty much saying that the risks went up pretty linearly with BP. So changed from 115 to 135 systolic would double your risk, going to 155 would double it again, and going to 175 would give you a 6x risk. Etc. The trend held when you broke it down into smaller quanta, so how many mmHg you considered significant depends on how much value you place on the increased risks. If you think 1.5x risk is significant, then a value of 10mmHg is something to worry about. Etc.

The JNC 7 guidelines have it mostly right from current evidence, though many people can argue that the huge secondary benefits of ACE I's make them a better first line drug than the thiazide-type diuretics. Honestly, the only reason thiazides were first line in those guidelines was because they were generic in 2003 when the guidelines came out. Since the ACEs are becoming generic, there's really no reason not to use them preferentially. Evidence also conflicts regarding what the appropriate blood pressure goals are for diabetics, as well as the best way to treat people in the "prehypertensive" range. JNC 8 will come out at some point and will hopefully address these and other concerns, but you can trust me that Na restricted diets will still be a recommendation.



Lol. Because a multitude randomized controlled trials of tens of thousands of people on statins that show decreases in mortality (from heart attack, stroke, as well as all cause mortality) doesn't mean anything right? Works better in some sub-populations than others, but personally, I think a decreased risk of death is a good thing. (Note: this is for reasonably dosed statins. High dose statins as being better for everybody was a theory that did not hold up, and medical practice has changed. There are some early studies that had increased risk of mortality, but newer statins at studied doses lower your risk of death substantially)

Even with *brand name* statins, cost effectiveness for *primary prevention* (not even secondary prevention) is <$25k per quality-adjusted life year gained, where even the UK's NIH uses ~$50k as the point where they call something cost effective.

Now, the effectiveness of all other non-statin cholesterol lowering medications on actually decreasing mortality is rather suspect, and many experts recommend that even if a statin alone isn't enough, using other medications to change the numbers doesn't seem to decrease mortality (or may even increase it). This is something still up for debate (as are high dose statins in certain higher risk populations), and is something that won't be clear until we have more studies about the non-statin drugs. (Funnily enough, the lipidologist in the video series linked above is apparently in the "treat aggressively with combinations of drugs" camp. At least in one of the later videos. He's welcome to that opinion, but it isn't the most evidence based one in the world)

Jeebus guys. I love how you are so quick to throw out real medical evidence justified by the highest levels of literature available because of some videos of a journalist/blogger you saw on youtube. Do some people exaggerate the benefits of things like a low Na diet or statin use? Sure. But to say that they aren't helpful? I hope that most of you are first or second years that haven't had any family or outpatient IM months yet, because otherwise, your professors have been doing you a real disservice.

Well played sir.

 

[jdh71]

?

There are studies out there backing up what I'm saying, I've posted
3 already. I don't understand your point here, jdh.

My point is largely related to how evidence based medicine get used and abused in what we do - EBM can only say what it can say - look very closely who gets excluded from any study, and who get included, yet we spend a lot of time extrapolating results generally. Some of the best cardiovacular studies ever done excluded women. This is especialy problematic for just about anything in the ICU

Any doctor that can be replaced by a computer . . . should be

 

[jdh71]

lmao. pretty sure it was a side jab my direction.

Eh. I've actually done some thinking about your computers should run the world ideas, and I don't hate them as much as I used to. AI could be very effectively used - though I still think there is some inherent danger there, but that's way out in left feild for this topic.

 

[pkwraith]

Yay, I love my delicious, delicious sodium.