FRANÇAIS HIV AND NUTRITION PREVIOUS PAGE NEXT PAGE SUBSCRIBE Chester Myers' Nutrition Series HIV/AIDS and Nutrition: Commentary re Proof-Type Studies
This draft originated with an article entitled "Is proof of efficacy needed in order to feed starving people" published in Canadian AIDS News, volume IX, Number 1 (summer, 1996). Chester Myers, July 1996
Introduction
I have often heard the comment "It's not been proven" as an excuse for not considering a new approach to an old problem, or for refusing to take action that might be life-saving. Nowhere is this approach more apparent than in the area of nutrition. I have concluded in many instances that the comment, "It's not been proven" may in fact mean "I don't know anything about it, and I don't want to become informed about it." An example of this is the argument that there is no "proof" that high levels of dietary sugar are unwise for those living with HIV/AIDS. While there is ample circumstantial evidence that high levels of dietary sugar may encourage oral candida infection, there have been no rigid studies that prove this. However, in the scientific literature there are data that describe Candida albicans growth characteristics on a variety of simple sugars (Deák & Novák, 1989). There is no evidence that, when in the human body, these yeast change their diet preferences away from these simple sugars. Furthermore, a large industry exists "because... [glucose, a monosaccharide] is unsurpassed as the carbohydrate source for yeast and other organisms" (monograph, CPC International Inc.). Logically, one can ask just what is to be gained by feeding high levels of dietary sugar to those living with HIV/AIDS. Would it not be better to err on the side of caution? If high levels of dietary sugars might encourage candida infection, and if they are unnecessary, then why take a chance? Considering there are also human studies (Sanchez et al, 1993) indicating that high levels of dietary sugar are immunosuppressive, one might consider a reluctance to counsel against high levels of dietary sugar as playing Russian Roulette with people's lives.
Waiting for proof-type arguments could be carried a step further to suggest that we not eat food until we see a study that proves we must eat to stay alive. Is there not a place for commonsense and logic in the above case regarding dietary sugar, as there surely is in deciding whether or not we should eat food?
In some cases, one makes the mistake of equating scientific study with proof approaches. While proof-type studies may indeed be scientific, they are by no means the only type of scientific study, and indeed, some proof-type studies can hardly be considered as scientific.
It would be convenient to fall back comfortably on a study that proves an issue. It might then appear that reasoning to a logical conclusion based on evidence is unnecessary. This is a dangerous assumption. The technician may be encouraged to rely only on proof approaches because s/he does not have sufficient knowledge to make an informed opinion based on sound scientific evidence. But, one must always examine just what has been proven even when a proof approach is taken. In addition, one must ask if the proven answer has practical relevance. It may not.
In the area of nutrition, proof-type studies are particularly prone to little practical relevance, and may even be misleading. It may be often possible to improve quality of life, including for people living with HIV/AIDS, without proof-type data. It is important, however, that any decision to use an unproven treatment be based on an understanding of the latest studies and background information from the scientific literature (not just a selected portion of what is available). One should also do everything possible to prevent harm. In making a decision about new treatments, we need to ask what "side of the fence" to argue from - should we use the treatment if indicated and unlikely to be harmful or refuse to use unless proven efficacious even when unlikely to do harm?
A variety of study methodologies provide evidence to support scientific hypotheses. Among the various types of studies, double-blind, placebo-controlled, prospective studies are considered a "gold-standard". Clinical trials to answer questions about the potential efficacy of treatments for diseases are often conducted in such a manner.
In the American Journal of Clinical Nutrition, Gladys Block, University of California at Berkeley, recently questioned the use of clinical trials (Block, 1995). Noting that they "give us rigorous answers to restricted questions", she commented that in addition to being limited in scope, they are "inappropriate vehicles by which to test hypotheses involving multifactorial diseases of long latency and with multiple, interacting nutrient effects." She specifically cautioned against using the infectious disease model of "one disease, one bug". The most compelling statement from Gladys Block is "What is sufficient, because it is necessary, is the use of our brains ... . the only answer lies in an intelligent synthesis."
In The American Journal of Gastroenterology, a group of doctors and researchers commented: "if one applied the criteria of double-blind randomized trials to all areas of medicine, one would find that a significant percent of our medical practice did not meet that requirement, since much of it is based on physiologic consideration rather than randomized trials." (Singer et al, 1992)
What is a "proof" approach?
The statement "It's not been proven" implies no study has proven the issue, that is, no double-blind, placebo-controlled study. In such studies, only one variable is normally examined at a time. If a drug is examined, then the study participants are randomly divided into two groups. One group receives the drug; the other group does not receive the drug but instead receives a look-alike placebo. Neither those participating in the study nor the researcher knows who gets the drug or who gets the placebo - both are "blinded". After the study is completed, it is unblinded in order to analyze the data.
What is an example of proof-type studies?
One proof-type study recently received a lot of attention. A Finnish study found no reduction in lung cancer among smokers after 5 - 8 years of supplementation with _-tocopherol or ß-carotene, and a possibly harmful effect on lung cancer by the ß-carotene. The authors also noted "this finding may well be due to chance." (ATBC, 1994) Even at conferences dealing with scientific matters, this study has been quoted at face value, and without qualification, as having shown that high levels of ß-carotene may be harmful.
There are several issues that make the results of this study of little practical significance.
(1) At least ten different carotenoids have been identified in human diets. Five have been implicated in human health. ß-carotene works in conjunction with vitamins E and C. Ingestion of high levels of only one carotenoid may impair absorption of others (Kostic et al, 1995; White et al, 1994). Can logical conclusions be drawn from studies where there is supplementation with only one carotenoid? Should not negative results have been expected because of likely depletion of other carotenoids? May there have been encouragement, or exacerbation, of an existing vitamin C deficiency in this already oxidant-stressed group of individuals?
(2) There are over 300 naturally occurring carotenoids. Beta(ß)-carotene is the best known, although lutein, lycopene and canthaxanthin are rapidly becoming rivals in this regard. Several may be converted to vitamin A in the body; this mainly occurring in the gut mucosal lining, but also occurring in the liver. The conversion requires the presence of vitamin E, and its activity requires at least zinc and iron (Hunt & Groff, 1990).
Smokers have been observed to have low zinc status (Faruque et al, 1995). The known requirement of zinc for vitamin A activity makes it likely that zinc deficiency would have been exacerbated in the Finnish study. However, this parameter was not monitored, and there is no recording that study participants were given a general multivitamin supplement to guard against such depletion of the one micronutrient as the result of administration of high levels of another single micronutrient (i.e., the vitamin E or the ß-carotene).
(3) "carotenoids with vitamin A activity include ß-apo-8'-carotenal, alpha-carotene, cryptoxanthin, echinenone, and torularhodin. .... The naturally occurring carotenoid-protein complexes apparently are more stable than the carotenoids per se. ....Powdered dehydrated fruits and vegetables have poor carotenoid stability unless stored in an inert atmosphere. .... Lycopene is perhaps the most unstable carotenoid. .... Canthaxanthin is somewhat more stable than beta-carotene, and beta-apo-8'-carotenal somewhat less stable, ..." (Borenstein & Bunnell, 1966).
In those studies showing possible negative results from carotene, only the synthetic form of ß-carotene has been used. Results from synthetic forms of vitamins E and K have been noted to vary from results observed for natural forms of these vitamins (Bendich, 1990; Hunt & Groff, 1990: RDA, 1989). Should unusual results have been expected when a synthetic form of that one carotenoid is used? For example, what were the levels of free radicals in the chemically synthesized ß-carotene supplements?
While there is still no validation of the theory that isolated naturally occurring carotenoids are as stable as in their natural habitats, it is generally noted "carotenes .. are often complexed with protein in food." (Hunt & Groff, 1990) Could the absence of an appropriate stabilizing compound have resulted in a high level of free radicals, or other species with potential health-harming effects?
(4) Finally, Hennekens (1994) suggested that "trials that last only a few years can assess protection only against the later stages of carcinogenesis. The lack of significant benefit ... does not necessarily preclude the finding of a significant benefit in substantially longer trials. ... the finding is so much at variance with the totality of other evidence suggesting a benefit".
It seems the Finnish study was an example of futile research. Indeed, at the recent 1996 Institute of Food Technologists (IFT) meeting, it was noted to be a sample of bad experimental design. Yet it was double-blinded and placebo-controlled. However, there was one useful item of information. In the placebo group, when the serum levels of alpha-tocopherol and beta carotene were examined, "the incidence of lung cancer was higher among the subjects in the lowest quartile group than among those in the higher." Thus, when obtained from natural sources, higher levels of both compounds were associated with benefit. However, there is no basis here to make any judgement regarding still higher levels.
But are other studies better?
It is not intended here to imply that proof-type studies are necessarily inappropriate. If well designed with regard to existing science they should be of considerable practical impact. The main objective of this commentary is to emphasize the importance of (i) having an informed hypothesis in hand before proceeding to the experimental protocol stage, and (ii) taking into consideration available relevant information at the time of interpretation of results, since during any study there may be publication of new data that need to be considered before making conclusions from the study at hand.
It is not intended here either to imply that other studies are not subject to problems in interpretation regarding practical use.
A study published in 1984 documented immune suppression among healthy young men who were given 300 mg of zinc per day over a 6-week period (Chandra, 1984). There was no documentation that copper levels were monitored, nor that copper supplements were given with the zinc supplements. At that time, no available data indicated issues regarding zinc/copper balance. Copper deficiency may have resulted from such zinc supplementation, and immune suppression or alteration would be an expected consequence of such copper deficiency (Chandra, 1991; Kelley et al, 1995). The author correctly pointed out that "changes in the levels of other nutrients also may occur and could contribute to the cellular effects observed in this study". He further stated that "such amounts are well tolerated and needed in patients with zinc malabsorption, eg, acrodermatitis enteropathica, Crohn's disease". While the study author correctly interpreted the data in light of available information, others have since used this 1984 study without referring to the author's note about malabsorption, and ignoring subsequent data on the zinc/copper issue or about improvements in health in the presence of zinc supplementation in HIV/AIDS.
... and relative to HIV/AIDS
In HIV disease, there are studies that ARE relevant and of practical significance. It is not intended here to review any of the many studies showing nutritional deficiencies such as vitamins, minerals, specific amino acids and secondary metabolites such as carnitine and glutathione in HIV/AIDS. A few comments are given here, however, to alert the reader to useful information that may be overlooked when there is a preoccupation with double-blind, placebo-controlled studies.
Information already given above as related to the Finnish study is relevant for people with HIV/AIDS since zinc deficiency is common with these people. Furthermore, vitamin A becomes deficient in HIV (Karter et al, 1995 and references cited elsewhere herein) and is associated with increased mortality (Semba et al, 1993). Vitamin A supplementation has been shown of benefit in children with HIV (Coutsoudis et al, 1995). Baum and colleagues from the University of Miami "because of the urgency" first recommended supplementation at the 1992 International AIDS Conference (Baum et al, 1992), and later recommended in a letter to the editor of AIDS for vitamin A supplements at 6-10 times the RDA to be included with a number of other vitamins also in supplemental form and this on the basis of "preliminary data from an ongoing nutrition supplementation trial" (Baum et al, 1994).
Five carotenoids known important for humans have been found to be low in people with HIV/AIDS (Coodley, 1994; Periquet et al, 1995). Relative to other micronutrients, the decrease in carotenoids has been noted to be "dramatic" (Sappey et al, 1994). Poor absorption is at least one source of this (Ullrich and co-workers, 1994, 1995). No clinical toxicity was observed in a study of 60 mg supplementation with ß-carotene in 11 patients with HIV over a 4-month period. Increases in percent natural killer cells were observed, and this effect was maximum at 3 months (Garewal et al, 1992). In another, controlled, study 180 mg/day of ß-carotene for 4 weeks produced no toxicity, statistically significant increases in white blood cell counts and statistically insignificant increases in CD4 counts (Coodley et al, 1993).
Similarly, with respect to zinc supplementation and the 1984 study cited above, apparently uninformed people have used the above cited 1984 publication as a blanket disapproval of supplementation with zinc at both the high levels used in the study and even amounts as low as 25 mg per day (Galvin, 1992; Holley et al, 1992), and this even in the presence of HIV/AIDS where zinc deficiency tends to be common, and malabsorption is also common. Perhaps with a potential for endangering the well-being of those trying to live with HIV/AIDS, the most important part of the publication for those with HIV/AIDS was overlooked. This was the statement by the study author noting the importance of very high levels of zinc supplements in cases of malabsorption such as in Crohn's disease. Surely a responsible approach here would have been to note that high levels of zinc supplementation need to be balanced with copper supplementation, and to warn that, since in HIV/AIDS malabsorption is a common problem, high levels of zinc may be necessary. Additionally, there have been other recommendations based on intervention study that levels of zinc above about 75 mg per day may be beneficial (Baum et al, 1994; Gordon, 1992, 1992; Mocchegiani et al, 1995, 1995a; Zazzo et al, 1989).
Again, this commentary is not an argument against proof-type studies, but rather a caution that (a) there are other types of study that produce useful data, and (b) it is important that special attention be paid to the design of proof-type studies in order to ensure that the narrow questions answered by such study are meaningful and hopefully useful. In the other extreme, there are even cases where opinion may be useful provided harm is unlikely to result. When it is necessary to resort to opinion, however, it is essential that such expression be informed opinion and that underlying information be both comprehensive and current.
In the area of nutrition, because of the interrelatedness of nutrients, meaningful study of single nutrients is difficult. With the establishment of accepted serum levels for most micronutrients, an accepted approach to deficiencies is to re-establish normal serum levels either by fortification of food or by supplementation. Accepted levels of intake are based on amounts required to achieve those serum levels considered normal. In the presence of HIV/AIDS, however, it has been shown that food alone cannot achieve such normal serum levels nor the benefits associated with such normal levels. In cases where calorie deficiencies are documented, an accepted approach has been to feed first. Maintaining starvation among starving people for purposes of collection of data from them is not considered ethical. Yet in HIV/AIDS treatment, we have the spectre of deficiencies and, in some cases, a refusal to correct these apparently because it is commonly accepted dogma in the absence of HIV/AIDS that food alone is enough. But, several studies show that food alone is not enough for those with HIV/AIDS; indeed, that eating more to attempt to get adequate micronutrients may have detrimental results. This may be because digestion of the macronutrients (protein, fat and carbohydrate) of our foods may cause a micronutrient deficit. Increasing the amount of the macronutrients may therefore increase the micronutrient deficit.
Supplementation has been shown to correct these deficiencies (references to Baum and co-workers). Improvements in both immune parameters and longevity have been shown to correlate with such supplementation (Tang et al, 1995; Abrams et al, 1993). These results have been observed in 6-year studies, each with about 300 people, making the results as dramatic as the beneficial results from any other approach currently used in the treatment of HIV/AIDS. Furthermore, the possibility that the beneficial results might have been from simply eating better was rigidly ruled out in both studies.
There are cases where proof approaches may be appropriate for the study of micronutrients, provided the study design ensures the unlikeliness of maintaining starvation. If therapeutic levels (i.e., levels well beyond those required for simple nutrition) of a particular micronutrient are of interest, then a double-blind, placebo-controlled approach would seem ethical provided both groups of the study receive sufficient supplementation to make deficiencies unlikely. An example here is vitamin E where an established modest level of supplementation has been shown to restore serum adequacy. Clinical studies would be useful to finetune appropriate levels, and furthermore to establish possible additional benefits beyond simple nutritional requirements by therapeutic levels such as 1500 IU. There are already indications that very high levels may provide benefits for the immune system (Alexander et al, 1985; Bendich, 1988, 1990b; Garewal et al, 1992; Murata et al, 1994). Again, the known health requirements for normal dietary levels and establishment of normal serum levels (references to Chandra; Peng et al, 1995; Pralhala et al, 1991) would make it unethical to do studies where one group is maintained at sub-normal levels.
In nutrition, the bottom line is we should not keep starving people starving. In the absence of supplementation, people living with HIV/AIDS are very likely to have multiple micronutrient deficiencies, and will therefore be in a state of micronutrient starvation (suffering from "hidden hunger"). Supplementation has been shown to correct this effect. Food alone has been shown as inadequate to correct it.
References
The Alpha-tocopherol, beta carotene cancer prevention study group (1994). The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. N Engl J Med. 330:1994:1029.
Alexander M and others, "Oral betacarotene increases the number of OKT4+ cells in human blood", Immunol Lett. 9:1985:221 (cited by Bendich 1990b).
Baum MK, "Dietary Intake Disturbances", International Conference Nutrition and HIV Infection, Cannes, 1995.
Baum MK and others, letter to the editor of AIDS (1994).
Baum MK and others, abstract PoB 3675, Amsterdam International AIDS Conference, 1992 (these initial recommendations were made on the basis of levels required to maintain serum adequacy).
Baum MK and others, "Micronutrients and HIV-1 disease progression", AIDS 9:1995:1051-1056.
Bendich A, "A Role for Carotenoids in Immune Function", Clinical Nutrition 7:1988:113-117.
Bendich A, "Antioxidant Nutrients and Immune Functions - Introduction", in Antioxidant Nutrients and Immune Functions, ed. A Bendich, M Phillips, RP Tengerdy, Series Advances in Experimental Medicine and Biology, Plenum Press, 1990.
Bendich A, "Antioxidant Vitamins and their Functions in Immune Responses", in Antioxidant Nutrients and Immune Functions, ed. A Bendich, M Phillips, RP Tengerdy, Series Advances in Experimental Medicine and Biology, Plenum Press, 1990b.
Block G, "Are clinical trials really the answer", Am J Clin Nutr. 62S:1995:1517S-1520S.
Borenstein B, Bunnell RH, "Carotenoids, Properties, Occurrence, and Utilization in Foods", Adv Food Res. 15:1966:195-276.
Chandra RK, "Excessive Intake of Zinc Impairs Immune Responses", JAMA 252:1984:1443-1446.
Chandra RK, "Cellular and Molecular Basis of Nutrition-Immunity Interactions", in Antioxidant Nutrients and Immune Functions, ed. A Bendich, M Phillips, RP Tengerdy, Series Advances in Experimental Medicine and Biology, Plenum Press, 1990.
Chandra RK, "1990 McCollum Award Lecture. Nutrtion and immunity: lessons from the past and new insights into the future", Am J Clin Nutr 53:1991:1087-1101.
Coodley GO (Oregon Health Sciences University, Portland, OR)"Vitamin and Mineral Update/Antioxidants", Third Internations Symposium on Nutrition and HIV/AIDS, Philadelphia, 1994.
Coodley and others, "ß-Carotene in HIV Infection", J Acquir Immun Deficien Syndr 6:1993:272-276.
Coutsoudis and others, "The effects of vitamin A supplementation on the morbidity of children born to HIV-infected women", Am J Public Health 85:1995:1076-1081.
Deák T and Novák DK, "Active and passive transport of monosaccharides in Candida albicans", and EK Novák, T Deák, "The role of transport process in the aerobic and anaerobic utilization of disaccharides by yeasts", Journal of Microbiology and Serology, Suppl. Yeast Symposium, Swets & Zeitlinger, Publishers, 1969.
Faruque MO and others, "Relationship between smoking and antioxidant nutrient status", Br J Nutr. 73:1995:625-632.
Galvin TAG, "Micronutrients: Implications in human immunodeficiency virus disease", Top Clin Nutr 7:1992:63-73.
Garewal HS and others, "A Preliminary Trial of Beta-Carotene in Subjects Infected with the Human Immunodeficiency Virus", J Nutr 122:1992:728-732.
Gordon AM de, "Effects of Adjuvant Therapy with Zinc in Human Immunodeficiency Infection", abstract 17, J Amer Coll Nutr 11:1992:601.
Gordon AM de, "Evaluation of Tissue Zinc Status through Cell-Mediated Immune Response in HIV Disease", abstract 18, J Amer Coll Nutr 11:1992a:602.
Hennekens CH and other, "Antioxidant vitamins - benefits not yet proved", N Engl J Med 330:1994:1080-1081.
Holley J and others, "Nutrition and HIV/AIDS: The Toronto Experience", conference for Health Care Professionals, 04 December, 1992, Toronto.
Hunt SM, Groff JL, Advanced Nutrition and Human Metabolism, West Publishing Co., 1990.
Karter DL and others, "Vitamin A deficiency in non-vitamin-supplemented patients with AIDS: A cross-sectional study", J Acquir Immun Defic Syndr Human Retrovir 8:1995;199-203.
Kelley DS and others, "Effects of low-copper diets on human immune response", Am J Clin Nutr 62:1995:412-416.
Kostic D and others, "Intestinal absorption, serum clearance, and interactions between lutein and ß-carotene when administered to human adults in separate or combined oral doses", Am J Clin Nutr 62:1995:604-610.
Mocchegiani E and others, "Benefit of oral zinc supplementation as an adjunct to zidovudine (AZT) therapy against opportunistic infections in AIDS", Internat J Immunopharm 17:1995:719-727.
Mocchegiani E and others, "Effect of zinc supplementation on opportunistic infections in AIDS", abstract P074, International Conference Nutrition and HIV Infection, Cannes, 1995a.
Murata T and others, "Effect of long-term administration of beta-carotene on lymphocyte subsets of humans", Am J Clin Nutr 60:1994:597-602.
Peng YM and others, "Concentrations and plasma-tissue-diet relationships of carotenoids, retinoids, and tocopherols in humans", Nutr Cancer 23:1995:233-246.
Periquet BA and others, "Micronutrient levels in HIV-1 infected children", AIDS 9:1995:887-893.
Pralhala R and others, "The effect of 13-cis-retinoic acid and beta-carotene on cellular immunity in humans", Cancer 67:1991:1556-1560.
RDA, Recommended Dietary Allowances, 10th ed., National Academy Press, 1989.
Sanchez A and others, "Role of sugars in human neutrophilic phagocytosis", Am J Clin Nutr 26:1973:1180-1184.
Sappey C and others, "Vitamin, trace element and peroxide status in HIV seropositive patients: Asymptomatic patients present a severe beta-carotene deficiency", Clin. Chim. Acta 230:1994:35-42.
Semba RD and others, "Increased Mortality Associated with Vitamin-A Deficiency during Human Immunodeficiency Virus Type-1 Infection", Arch. Intern. Med. 153:1993:2149-2154 (also see works by same group in Clin Infect Dis 21:1995:966-972; Lancet 343:1994:1593-1597).
Singer P, Katz DP, Dillon L, Kirvelä O, Lazarus T, Askanazi J, "Nutritional Aspects of the Acquired Immunodeficiency Syndrome", The American Journal of Gastroenterology 87:1992:265-273.
Tang A and others, "The Effect of Micronutrient Intake on Survival in HIV-1 Infection", presentation by N Graham, Toronto, 1995.
Ullrich R, "HIV Enteropathy", International Conference Nutrition and HIV Infection, Cannes, 1995.
Ullrich and others, "Serum Carotene Deficiency in HIV-Infected Patients", AIDS 8:1994:661-665.
White WS and others, "Pharmokinetics of beta-carotene and canthaxanthin after individual and combined doses by human subjects", J Am Coll Nutr 13:1994:665-671.
Zazzo JF and others, "Effect of Zinc on the Immune Status of Zinc-depleted AIDS Related Complex Patients", Clin Nutr 8:1989:259-261. |
