expert opinion

The complexity of vitamin C research

July 1, 2014


Alexander J. Michels, PhD, Linus Pauling Institute, Oregon State University, Corvallis, USA

“Ascorbic acid, the reduced form of vitamin C, is an essential component of the human diet. Small amounts of ascorbic acid can prevent the deficiency disease, scurvy, while accumulation of high levels of ascorbate in plasma and tissues may protect against oxidative damage and limit inflammation. However, ascorbic acid is unlike many other vitamins owing in part to its unique redox chemistry. In addition, the tight control of vitamin C status and metabolism in the body, along with biological effects of supple- mentation that may differ between animals that can synthesize ascorbate versus species that cannot syn- thesize ascorbate, like humans, set it apart in the micronutrient field. Thus, many common research practi- ces that are sufficient for the study of other vitamins and minerals are often inadequate for the study of vitamin C, leaving the specific challenges to the design and execution of experiments utilizing ascorbic acid underappreciated.

Currently, studies involving vitamin C consumption in human subjects are not held to a rigorous standard. Many, but not all, prospective cohort studies have observed associations between increased vitamin C intake or plasma levels and a decreased risk in developing chronic diseases, including coronary heart disease, ischemic strokehypertension, and certain types of cancer (1). However, several large randomized cont- rolled trials (RCTs) have shown no benefit of vitamin C supplementation when taken alone or in combination with other micronutrients (2). This apparent failure of vitamin C supplements to affect human health can be attributed to many factors related to study design. The most predominant is the use of the standard RCT study design, which is intended to test the safety and efficacy of a pharmaceutical drug in individuals that are at high risk or are suffering from a condition or illness. By contrast, enrollees in vitamin C supplementation studies, and diet-related RCTs in general, are usually health-conscious individuals who are likely to consume an above-average diet and maintain a healthy body weight. As a consequence, these individuals have a lower disease incidence and a better nutritional status, including vitamin C, than the general population – both of which negatively affect the statistical power of the study. Statistical power is further compromised by the fact that there is no true placebo group in these RCTs, as even the non-supplemented subjects continue to obtain vitamin C from their diet throughout the duration. These and other serious flaws in study design, including lack of a single supplement (vitamin C only), quality of the methodology employed, and lack of discrimination by genetic polymorphisms, have led some to the unfortunate conclusion that very few well- designed, well-controlled trials of supplemental vitamin C have ever been conducted (2).

One reason previous studies have failed to show health benefits of vitamin C may be the assumption that an individual’s plasma or body ascorbate status directly reflects their dietary or supplemental intake of vitamin C. To the contrary, analysis of food frequency questionnaires has revealed that there is little correlation between asses- sed vitamin C intake and plasma ascorbate levels, likely due to inaccuracies in dietary assessment methodology using food frequency questionnaires or food diaries, loss of ascorbate during storage, cooking or processing, and large inter-individual differences in vitamin C absorption and metabo- lism. An example of the latter is the lower plasma ascorbate levels observed in the elderly when compared to younger adults consuming equivalent amounts of vitamin C, suggesting changes in absorptive capacity with age. In addition, smoking, chronic aspirin use, high alcohol consumption, high BMI, and low socioeco- nomic status are all factors that have been associated with lower plasma vitamin C levels. Furthermore, genetic variations linked to vitamin C metabolism also may lead to altered plasma ascorbate levels depen- ding on the various polymorphisms involved. Thus, food frequency questionnaires have little predictive value for evaluating the effect vitamin C consumption on disease risk, while plasma ascorbate levels display a clear relationship. Plasma concentrations start leveling off at doses above 200 mg/day and approach maxi- mal levels in the range of 60–90 μM.

From the above considerations, three critical issues emerge in relation to RCT design. First, individuals recruited for a research study should have low plasma ascorbate levels at baseline to increase the likelihood of affecting changes in ascorbate status in tissues through vitamin C supplementation. Subjects already con- suming enough vitamin C to provide near-maximal or saturating plasma and tissue levels of ascorbate are highly unlikely to demonstrate any further biological or health effects upon vitamin C supplementation. Second, the intervention must be proven effective, demonstrating – at the very least – an elevation in plasma ascorbate steady-state levels. Again, if a research subject’s vitamin C status does not change, no changes in health or disease outcomes can be expected unless it can be supported by an alternate mecha- nism. In many cases, no biological effect can be expected of increasing vitamin C levels if no functional deficit is present. For instance, although studies support the use of vitamin C in improving vascular function and reducing blood pressure (3), continued supplementation of vitamin C when plasma levels are already at saturation will not yield additional vasodilation, and changes in blood pressure are not expected if a subject is already within a healthy blood pressure range. Lastly, in the absence of tissue ascorbate measurements, the study design and endpoints must relate to our knowledge about the distribution of vitamin C in the body. For example, if brain ascorbate levels are near saturation at low vitamin C intake and plasma levels, it is unrea- sonable to expect an effect on brain function over a wide range of intake and plasma ascorbate levels.

Many excellent evidence-based reviews have summarized the health effects of vitamin C, focusing on vita- min C’s possible role in the prevention or treatment of cardiovascular disease, cancer, diabetes, and other diseases (1, 2, 4). Despite the large number of research studies, the evidence supporting the effects of vitamin C supplements on common cold incidence and duration has been relatively weak, with the exception of marathon runners, skiers, soldiers in subarctic conditions, or individuals with chronic gastritis (5). It cannot be overemphasized that the primary role of ascorbic acid in biological systems is that of a reductant, and the most established health effects of this reductive power are related to ascorbate’s role as an electron-dona- ting enzyme cofactor used, e.g., for pro-collagen, carnitine, and catecholamine biosynthesis (6). Ascorbic acid has many additional roles in the body beyond these enzyme functions, for example antioxidant protec- tion, and likely more roles will be elucidated in the future. With a critical evaluation of all potential roles of ascorbic acid, the borders of vitamin C research can advance with balanced, evidence-based approaches. Already, emerging roles for vitamin C in various hydroxylase enzymes have recently placed a focus on gene expression and changes in histone and DNA methylation, suggesting vitamin C may regulate global changes in gene expression. Further innovation has been demonstrated in studies on vitamin C bioavailability, genetic variation of the vitamin C transporters, and intravenous vitamin C infusions in cancer therapy (7).

As vitamin C research progresses into the 21st century, it has become clear that much more work still lies ahead of us. The experimental faults, artifacts, and myths currently afflicting vitamin C research limit the impact of many studies, making their contributions to general knowledge of the biological roles of ascorbic acid unremarkable at best, and confusing and detracting from the real issues at worst. This can also stretch beyond the realm of laboratory research, as the persistence of poorly controlled studies within this field often undermines efforts in the medical community to recommend vitamin C as a safe, effective means of promo- ting health. If nothing else, it weakens efforts to fund additional, well-designed RCTs necessary to establish definitive health claims that are desperately needed. In light of these issues, we must increase scrutiny of vitamin C studies in the past as well as the present, holding them to a higher standard based on the evidence discussed here if we are to make a lasting contribution to our understanding of vitamin C’s impact on human health.”

Based on: Michels A. J. and Frei B. Myths, Artifacts, and Fatal Flaws: Identifying Limitations and Opportunities in Vitamin C Research. Nutrients. 2013; 5:5161–5192.

1. Frei B. et al. Authors’ perspective: What is the optimum intake of vitamin C in humans? Crit. Rev. Food Sci. Nutr. 2012; 52:815–829.

2. Lykkesfeldt J. and Poulsen H. E. Is vitamin C supplementation beneficial? Lessons learned from randomized controlled trials. Br. J. Nutr. 2010; 103:1251–1259.

3. Juraschek S. P. et al.. Effects of vitamin C supplementation on blood pressure: A meta-analysis of randomized controlled trials. Am. J. Clin. Nutr. 2012; 95:1079–1088.

4. Tveden-Nyborg P. and Lykkesfeldt J. Does Vitamin C deficiency increase lifestyle-associated vascular disease progression? Evidence based on experimental and clinical studies. Antioxid Redox Signal. 2013; 19(17):2084–2104.

5. Hemila H. and Chalker E. Vitamin C for preventing and treating the common cold. Cochrane Database Syst. Rev. 2013; 1:CD000980.

6. Michels A. J. and Frei, B. Vitamin C. In Biochemical, Physiological, and Molecular Aspects of Human Nutrition, 3rd ed.; Stipanuk, M.H., Caudill, M.A., Eds.; Elsevier/Saunders: St. Louis, MO, USA, 2012; pp. 626–654.

7. Welsh J. L. et al. Pharmacological ascorbate with gemcitabine for the control of metastatic and node-positive pancreatic cancer (PACMAN): Results from a phase I clinical trial. Cancer Chemother. Pharmacol. 2013; 71:765–775.