Subject: I review the published studies regarding a multi-level product called Juice Plus.
Someone sent me an e-mail about Juice Plus, a multi-level-company he's involved with. Typically I have a simple approach to multi-level evangelists, “Don't believe a word they say.”
I replied that I hadn't seen much information published about this product and wasn't inclined to place much credence on self published papers. I came into work a few days later to find a large file of glossy reprints of numerous studies on my desk.
I leafed through the pile and was impressed. Enough so that I thought I should rethink my earlier opinions and debated how best to start referring patients to purchase these products.
That weekend I came home intending to write a newsletter summarizing the studies I'd been reading but forgot to bring the file. “No problem I thought,” I can do a quick search on PubMed and find them all. I could only find one trying every search heading I could. So I went to the company's website where they had a long list of studies accompanied by claims on how they were the most researched nutritional supplement in the universe (I exaggerate but you get my point).
Struggling to keep an open mind I started looking more closely. Many of the studies I had read at the office turned out had not been published in peer review journals. Instead they had been presented as abstracts (a process that doesn't entail peer review) at various professional conferences. The company selling the product printed the studies up in glossy booklets. I should have known better and missed the small print.
I went through the list of studies on the Juice Plus website and did find several which had been published. I looked them up and then looked at related studies so I could have some context to judge them by.
The studies come up lacking. I will talk about the questions I have and discrepancies in a moment. First though, I must admit that I do not need studies to believe that something works. Nor does the lack of good studies make me think something is ineffective. Heck, look at prune juice. It was only a few years ago that someone published the news that it had a laxative effect. Yet when a company relies heavily on “scientific” studies to promote their product, I expect them to do so.
Over the years I have heard from many patients that this product has been beneficial for them. In fact I am hoping this newsletter provokes a great many testimonials to that effect which I will be both happy to receive, to look at and to pass along in due time.
There are several studies on Juice Plus published in peer reviewed journal over the last 7 years. I want to look at these studies individually and compare them against similar studies. I am not going to consider the unpublished abstracts which the company uses to promote their product.
A study done in Australia , published in July, 2003, says Juice Plus (JP) lowers homocysteine levels  . Since homocysteine levels are directly correlated with increasing heart disease risk anything that lowers homocysteine seems attractive. Two weeks of Juice plus and “Plasma homocysteine was reduced (8.2 +/- 1.5 vs. 7.6 +/- 1.1; P < 0.05)” This was a placebo controlled crossover study so what those numbers mean is that after taking the Juice Plus participants' homocysteine levels averaged 7.6 versus 8.2 before they took it. Looking at the confidence intervals there was considerable overlap; the JP group was somewhere in the 6.5 to 8.7 range and the placebo group was 6.7 to 9.7. In other words these values are not that different.
What other things lower homocysteine and by how much?
In a study which gave a mix of the standard nutrients associated with lowering homocysteine (0.65 mg folic acid, 150 mg alpha-tocopherol, 150 mg ascorbic acid, 12.5 mg beta-carotene, and 0.4 microgram vitamin B12) for 15 days, “Serum homocysteine levels decreased from 8.7 +/- 4.3 to 6.3 +/- 2.2”  Compared to the Juice Plus study which lowered homocysteine by ~0.6, this vitamin mix lowered it by 2.4, four times as much. The truth is that these numbers can not be compared. The P value for these two studies was different. This vitamin study used a P value of .001 versus a P value of .005 for the JP study. The numbers specified with the lower P value are more accurate. It's been awhile since I took statistics but my guess is that if the JP study data was expressed statistically at a P=.001 level the before and after results would have been statistically insignificant.
There's another study which looked at dietary intervention instead of pills. In this study, subjects started by eating only low folate foods and then “ subjects ate at least seven servings of vegetables, berries, and citrus fruit/d.” for a period of five weeks.  The mean plasma homocysteine concentration was 8.0 pmol/ at the end of the low-folate diet. This is interesting as even after the low folate stage, the subjects had lower homocysteine levels than the Juice Plus participants suggesting that they too may have been folate deprived at the start of the study. After the five week period of high folate diet, homocysteine decreased by 13 % (P<0.001). That would be a drop of about 1.0. The dietary effect was thus larger and statistically more significant (P=.001) than the JP study.
In another study looking at the effect of five a day of fruits and vegetables researchers compared diets which contained about half a kilo (one pound) of fruits and vegetables a day against one which contained about 100 grams (about 4 ounces). The five a day diet lowered homocysteine by about 11% in 4 weeks. I am unable to tell from the abstract the homocysteine level of the “low” folate dieters. The 4 ouncers may have still had lower homocysteine than the controls in some of these other studies. 
Another supplement suggested for lowering homocysteine levels is betaine. There's a study comparing betaine's effect against folic acid. Participants consumed either a placebo or 6 grams of betaine, or betaine and folic acid (800 mcg/day) for 6 weeks. Betaine lowered homocysteine levels by 1.8, while folate knocked it down by 2.7. Clearly folic acid worked better. 
I would like to compare the homocysteine lowering effect of Juice Plus against these other interventions but it is difficult to do so. The values calculated for Juice Plus used a weaker statistical P value than the other studies so we can be less sure of the numbers. If we ignore the differing P values and just look at the end results as if they were computed in a comparable manner, JP lowered homocysteine by 0.6 while all the other interventions lowered it by significantly more. Dietary changes lowered it by about 1.0 or almost twice as much. Combination vitamins lowered it by 2.4. Betaine alone lowered homocysteine by 1.8 and folate alone by 2.7.
A second Juice Plus abstract which was published in a peer review journal:
Plotnick, G.D., Corretti, M.C., Vogel, R.A., et al., "Effect of Supplemental Phytonutrients on Impairment of the Flow-Mediated Brachial Artery Vasoactivity After a Single High-Fat Meal," Journal of the American College of Cardiology, 41(10), 2003, pages 1744-1749.
This one is a complicated to explain. Let me summarize this study and then look at it more carefully. Eating a high fat meal has an effect on the brachial artery. It is unable to dilate normally for about four hours. This phenomenon has been formalized into a standardized test, the flow-mediated brachial artery reactivity test (BART). In this study taking Juice Plus for four weeks blunted the effect of a high fat meal from a 45% decrease in vasoreactivity to only 7%. Put simply, high fat meals cause nasty to happen in our arteries and Juice Plus decreases the extent of the nasty effect.
Let me back up and look at this more carefully. This test was developed to help explain the development of atherosclerotic disease. Running the various cholesterol levels has not been as predictive as researchers would have hoped. There are national variations and individual exemptions from the basic predictive models for disease risk. This goes back to the French Paradox. There are not just people, but whole nations of people, who have crummy cholesterol levels but who don't develop the predicted atherosclerosis. In the BART, subjects are fed a defined high fat meal: 900 calories including 50 grams of fat. Their brachial arteries are occluded (squished shut) using a blood pressure cuff for five minutes. Then one minute after the pressure on the artery is released, its diameter is measured using an ultrasound. Typically after this squishing procedure the artery dilates, that is opens up, probably to deliver more blood to the oxygen starved limb. No change in this vasodilation is seen after a low fat meal. Yet after a high fat meal it happens. The artery isn't able to open up as wide. In one earlier study, flow-dependent vasoactivity decreased from 21 % before eating to 11 %, 11 %, and 10 % at 2, 3, and 4 hours after the high-fat meal, respectively.  Sorry about all the numbers but they will get important in a moment. Still in simple words, the arteries dilated only about half as much as they normally did.
A number of interventions have been studied which prevents this fat effect. Yale researchers have shown that either a bowl of oatmeal or an 800 iu capsule of vitamin E prevents the fat's effect on vasoactivity completely.  The study mentions that wheat cereal didn't prevent the fat effect. A bowl of wheat cereal according the study did not change the decrease in vasoactivity from baseline but this is interesting, their baseline was a 13% decrease. Back to Australia for a second, their baseline decrease was 45% and their Juice Plus treated decrease was 17%. Something seems wrong here.
Do Australians start with a base line fat induced vasoreactivity decrease that is 2 to 3 times as high as the baseline established at Yale?
A study at Boston University was unable to show any effect of drinking red wine on BARTs. Though a slightly different study model, they dosed fat at 0.8 grams per kilogram body weight rather than the 50 grams for all participants; their baseline 10% decrease seen in BARTs is much closer to the Yale values than the Australian values.
A Chinese group has tested the effect of Vitamin C on this phenomenon. A two gram dose of ascorbic acid also stops the fat induced change completely. 
In looking at these studies and comparing them to the Juice Plus study several points come up. The first is the discrepancy in effect of the high fat meal. The Juice Plus study reports a baseline effect of close to 50% decrease in vasoreactivity after the fat meal. Other studies measure the baseline untreated effect at about 10 to 15%. The protocol followed does not appear different and I have not found an explanation. Remember the Juice Plus intervention changed the decrease in vasoreactivity from 47% to 17%. But this great end result is close to the baseline seen in other studies. In other words, the participants who took Juice Plus for a month ended up with test results that looked like the starting results seen in other studies.
The second point to consider is that a number of other interventions erase the fat effect completely. Vitamin E, oatmeal and vitamin are three examples. These three were single dose interventions taken with the high fat meal. Juice Plus was taken for four weeks and only reduced the effect. A bowl of oatmeal completely cancels out the effect of the fat on blood vessels. Though one can't always have a bowl of oatmeal, taking vitamin E or Vitamin C with a high fat meal is simple and works as well as the oatmeal.
 Samman S, Sivarajah G, Man JC, Ahmad ZI, Petocz P, Caterson ID.
A mixed fruit and vegetable concentrate increases plasma antioxidant vitamins and folate and lowers plasma homocysteine in men. J Nutr. 2003 Jul;133(7):2188-93.
 Nutrition. 2000 Feb;16(2):107-10. Related Articles, Links
Effects of supplementation with folic acid and antioxidant vitamins on homocysteine levels and LDL oxidation in coronary patients.
Bunout D, Garrido A, Suazo M, Kauffman R, Venegas P, de la Maza P, Petermann M, Hirsch S.
 Br J Nutr. 2003 Mar;89(3):295-301. Related Articles, Links
Plasma homocysteine concentration is decreased by dietary intervention.
Silaste ML, Rantala M, Alfthan G, Aro A, Kesaniemi YA.
 Broekmans WM, Klopping-Ketelaars IA, Schuurman CR, Verhagen H, van den Berg H, Kok FJ, van Poppel G. Fruits and vegetables increase plasma carotenoids and vitamins and decrease homocysteine in humans. J Nutr. 2000 Jun;130(6):1578-83.
 J Nutr. 2003 May;133(5):1291-5.
Betaine supplementation lowers plasma homocysteine in healthy men and women.
Steenge GR, Verhoef P, Katan MB .
 Vogel RA, Corretti MC, Plotnick GD. Effect of a single high-fat meal on endothelial function in healthy subjects. Am J Cardiol. 1997 Feb 1;79(3):350-4
 Katz DL, Nawaz H, Boukhalil J, Giannamore V, Chan W, Ahmadi R, Sarrel PM Acute effects of oats and vitamin E on endothelial responses to ingested fat.Am J Prev Med. 2001 Feb;20(2):124-9.
 Ling L, Zhao SP, Gao M, Zhou QC , Li YL, Xia B Vitamin C preserves endothelial function in patients with coronary heart disease after a high-fat meal. Clin Cardiol. 2002 May;25(5):219-24