Clinical Studies With Food Highlight Limitations of Nutrition Research
Today, two studies that conducted clinical trials using food appeared on my nutrition research radar: a study by Navarro and co-workers looking at the effect of cruciferous vegetables on markers of inflammation, and another by Pereira-Caro and colleagues on the bioavailability of polyphenols from orange juice.
To summarize the results briefly, the study in cruciferous vegetables supplemented 32 healthy men with four diets containing varying levels of cruciferous vegetables for two weeks each in cross-over fashion . Cruciferous vegetables are from the cabbage family and include broccoli, cauliflower and radishes. They are under study because the isothiocyanates that they contain show anti-cancer properties in the lab, and greater consumption is associated with reduced risk of cancer (see review by Higdon at al.). Diets were either vegetable-free, “single dose cruciferous” containing 7 g per kg bodyweight per day cruciferous vegetables (for someone weighing 70 kg or 155 pounds, this would be almost 500g per day!), “double dose cruciferous” containing 14 g per kg body weight per day, and “cruciferous plus apiaceous” with 7 g cruciferous and 4 g apiaceous vegetables per kg body per day. Apiaceous vegetables include carrots, celery, parsley, dill and parsnips. The researchers found significant decreases in circulating levels of systemic inflammation marker IL-6 in the “double-dose cruciferous” and “cruciferous plus apiaceous” diet groups. High levels of this marker are implicated in many diseases related to inflammation such as cancer, diabetes and heart disease. Lowering levels may therefore prevent these chronic diseases.
The orange juice study looked at the bioavailability of various metabolites from orange juice with extra pulp in a group of 12 healthy men and women. High consumption of oranges, orange juice or other citrus fruits is associated with reduced risk of stroke in women (Cassidy and co-workers), and studies with individual flavanones from oranges improves measures of cardiovascular function. The authors identified 8 flavanones in the orange juice, and these were converted to 14 urinary metabolites after consumption. Around 16% of the flavanones were excreted directly as metabolites, while the rest were consumed by the gut microflora, absorbed and excreted in the urine.
These studies both highlight the complexity of our food. Cruciferous vegetables contain a dizzying array of bioactive components (Ishida and associates), levels of which may vary according to species, cultivar, growing conditions, harvest conditions, part of the plant eaten, and also preparation and cooking. The participants in the vegetable study increased their consumption of a cocktail of phytonutrients, which can be replicated only in a broad sense. Likewise with orange juice: whereas some people may compare it to soft drinks, oranges contain a wide range of antioxidants that are also retained after processing steps such as squeezing, concentrating, freezing and pasteurization used to produce commercial orange juices. Quite simply, orange juice contains more than just sugar (see report from Gil-Izquierdo, Gil and Ferreres). The specific contents are variable however and no two glasses of orange juice could be expected to have the same mix of polyphenols.
While both studies used a randomized, controlled design, as is done with drug studies, the “investigational product” is in fact undefined except in a general sense. One reason why nutritional studies cannot be compared to drug studies is that for essential nutrients, we cannot use a true placebo group. Another reason is that foods contain an astonishing number of different chemicals. Even if we find associations between a certain group of vegetables, phytonutrient or type of food from research and a health effect, the exact components present that produce the health benefit may be difficult or impossible to isolate. This is due to the complex composition of foods, and potential interactions between nutrients, genetic factors and many other considerations.
This is one of the reasons behind key nutrition experts’ reservations about the quality of nutrition research, as described by Moyer earlier this year in Nature. It is difficult enough to show that supplemental vitamins improve health, even when we can put them in a pill, due to differences in baseline status, individual response and difficulties assessing normal intakes from food. The current studies also show how complex our foods are. While I would prefer to take a “broccoli pill” to reduce my risk of Alzheimer’s disease than consume roughly 800 g cruciferous vegetables every day (my requirement based on my body weight for the “double dose”), the problems with translating food-based research to clinically meaningful results make proving this concept very difficult.
Sandi L. Navarro, Yvonne Schwarz, Xiaoling Song, Ching-Yun Wang, Chu Chen, Sabrina P. Trudo, Alan R. Kristal, Mario Kratz, David L. Eaton, and Johanna W. Lampe. Cruciferous Vegetables Had Variable Effects on Biomarkers of Systemic Inflammation in a Randomized Controlled Trial in Healthy Young Adults. J. Nutr. jn.114.197434; first published online August 27, 2014. doi:10.3945/jn.114.197434
Gema Pereira-Caro, Gina Borges, Justin van der Hooft, Michael N Clifford, Daniele Del Rio, Michael EJ Lean, Susan A Roberts, Michele B Kellerhals, and Alan Crozier. Orange juice (poly)phenols are highly bioavailable in humans. Am J Clin Nutr 2014 ajcn.090282; First published online August 27, 2014. doi:10.3945/ajcn.114.090282
Cassidy A, Rimm EB, O'Reilly EJ, Logroscino G, Kay C, Chiuve SE, Rexrode KM. Dietary flavonoids and risk of stroke in women. Stroke. 2012 Apr;43(4):946-51. doi: 10.1161/STROKEAHA.111.637835. Epub 2012 Feb 23. http://www.ncbi.nlm.nih.gov/pubmed/22363060
Gil-Izquierdo A, Gil MI, Ferreres F. Effect of processing techniques at industrial scale on orange juice antioxidant and beneficial health compounds. J Agric Food Chem. 2002 Aug 28;50(18):5107-14. http://www.ncbi.nlm.nih.gov/pubmed/12188615
Higdon JV, Delage B, Williams DE, Dashwood RH. Cruciferous vegetables and human cancer risk: epidemiologic evidence and mechanistic basis. Pharmacol Res. 2007 Mar;55(3):224-36. Epub 2007 Jan 25. http://www.ncbi.nlm.nih.gov/pubmed/17317210
Moyer, Melinda Wenner. Nutrition: Vitamins on Trial. Nature News Feature. Published 25 June 2014. http://www.nature.com/news/nutrition-vitamins-on-trial-1.15459
Rapisarda P, Tomaino A, Lo Cascio R, Bonina F, De Pasquale A, Saija A. Antioxidant effectiveness as influenced by phenolic content of fresh orange juices. J Agric Food Chem. 1999 Nov;47(11):4718-23. http://www.ncbi.nlm.nih.gov/pubmed/10552879