Date: 24 Aug 1994 09:31:36 -0700 From: taltar@vertigo.helix.net (Ted Altar) Subject: garlic & cholesterol (repost) Newsgroups: rec.food.veg Organization: Helix Internet Lines: 244 GARLIC FOR LOWERING BLOOD CHOLESTEROL? Ted Altar Is garlic good for you? This is not as an easy question as it might appear. By "good for you" we either refer to its composition of dietary nutrients, or to its much touted effect on blood pressure and blood cholesterol levels, coagulation, platelet aggregation and vasodilation. With respect to the first "good for you" there don't appear to be that much in the way of vitamins and minerals. Consider the following table of a 100 grams of raw, peeled garlic cloves. % Of RDA For Adult Male (25-50 years) Obtain From Eating 100 Grams Of Raw Garlic vit C 28 vit A 0% B6 19 vit D 0% thiamin 9 carotene 0% niacin 6 B12 0% folate 3 pantothenate 0% riboflavin 2 biotin 0% vit E .1 phosphorus 21% iron 19% magnesium 7% zinc 7% calcium 5% selenium 3% iodine 2% As you can see, not a great range or quantity of the above nutrients are to be found in garlic. A 100 g. quantity of garlic, however, does have more than twice as much protein as 100 g of milk (7.7g vs 3.2) And it certainly does have more fiber than milk (4.1 g vs 0) and less fat (6% vs 53% fat by total calories). The only vitamins and minerals that garlic contains in any appreciable quantities are C, B6, phosphorus and iron. The later two, being minerals, should not be affected by the processing of raw garlic into garlic powder. If heat is used in the drying process, then 50% or more of the vit C could be lost by destruction due to heat and oxidation. In milling wheat, up to a 70 to 90% loss of B6 can occur, and I guess we could expect something similar here as well as garlic is dried and ground up. However, I really don't know for certain. Maybe if someone has tables for both raw and powdered garlic, then we could be more certain of exactly how much of a nutrient loss occurs with the dried, powdered variety. Fresh, raw garlic does have more nutrients, at least with respect to vitamin C and B6 that are present in garlic in nutritionally useful amounts. Now, the other meaning of "good for you" is harder to answer, partly because the research is not that definite regarding those CHD risk reduction benefits. The active component is thought to be ALLICIN, a sulphur containing compound, which, with its breakdown products, causes that malodorous bouquet that we are all familiar with. ;-( There are other biologically active compounds related to allicin, like ajoene, which are also present. Now, there is evidence that extracts or oils of garlic prepared by using steam distillation or organic solvents may have little activity (1). Carefully dried sliced cloves, however, do retain their potency of allicin. Maybe the same can be said about dried garlic powder, but it does seem unlikely that a commercial industry making garlic powder for seasoning would show the same due care. So, maybe there would be some loss, but not as much as one sees with the garlic extracts or oils. Another reason to not waste money on those silly food extracts in a pill but instead buy whole foods. An important consideration here is also remember that allicin in garlic can naturally vary between plants by a factor of 10! Part of the reason, some would argue, for the inconsistent empirical results of garlic on health would be these problems in ensuring that studies used active products and that the amount allicin was properly quantified. One review (3) did conclude that large does of fresh garlic (7-28 cloves a day!) did have a beneficial effect on cardiovascular risk, but that there is no convincing evidence that the commercial preparations had a comparable benefit. So again, save your money and don't buy those concentrated garlic capsules. Now, permit me to report the results from one German double- blind, four month study that was well controlled. This was mentioned by Mansell & Reckless (2). 261 German patients with high serum cholesterol and triglyceride levels were used. Therefore, some caution here should be exercised as to whether or not a similar effect is going to occur with normal, healthy adults like yourself. Results Of Taking 800 Mg Dried Garlic Powder/day before after total cholesterol (mmol/l) 6.87 6.07 triglycerides (mmol/l) 2.55 2.12 blood pressure (mm Hg) 171/102 152/89 This 13% drop in total cholesterol over 4 months is a pretty good result, although other studies usually get smaller drops of about 5%. But even this result is not so great when we compare how diet alone, at no added cost, will drop cholesterol leves by 25% in a mere 3 weeks (7) As to whether the double blind was successful (most people can detect having eaten normal garlic preparations), the authors report that the smell of garlic was reported (most frequently by spouses) in 21% of the treated patients and 9% of those taking a placebo. Evidence that garlic reduces cardiovascular risk is accumulating, but is still incomplete. It should also be noted that even onions (9) give some of the same effects, like the antiplatelet (4) effect noted above by blocking thromboxane synthesis for several hours. Also, the problem of an odorless garlic preparation to used in a double-blind study has not been satisfactory resolved. A preparation with no odor may also have no allicin. Although active preparation itself *can* be odorless, upon ingestion there then can be a good chance of a detectable odor occurring, as the above results indicates. Whether the amounts of garlic normally eaten in the household is going to be beneficial is really not very evident I think you would need to eat about 1 or 2 cloves of garlic to obtain the levels of allicin needed for some of the reported effects, but given the variability of allicin among garlic cloves, this would be hard to determine (remember, the dose is the medicine). Like aspirin, garlic can also "thin" the blood and have an antiplatelet effect, but too much can make for some undesirable side effects, such as spinal haematoma. Some uncommon but side effects include gastrointestinal disturbance, asthma, and contact dermatitis. One `in vitro' (in test tubes) experimental study (5) it was demonstrated that garlic does have a mild antibiotic activity against a broad spectrum of 17 strains of fungi, and was more effective against pathogenic yeasts than nystatin. What this would mean `in vivo' (in the body) I don't know. Some other natural foods for possibly reducing risk for CHD include: - ginger (10) for inhibit platelet aggregation - 20 gm of brewers yeast daily (11) for lowering LDL while raising HDL cholesterol. Presumably, nutritional yeast would work just as well. - 1/2 cup of beans daily (12) for lowering LDL - 10 tsp. alfalfa seed powder daily redcue total blood cholesterol levels by as much as 20% and improved HDL/LDL rations by up to 40% (13) - 300 mg daily of phytosterols (cholestrol analogs found in vegetables) lower serum cholesterol and may decrease cholestrol absorption. (14). - soy lecithin lowered LDL and triglycerides while raising HDL cholesterol (15) - bromelain, a proteolytic enzyme from pineapples, is thought to inhibit platelet aggregation (16) In sum, I don't think eating 1 or 2 cloves of garlic will do any harm, and it may well do some good. Of course, there is the indeterminate harm to one's social life that may occur from eating so much garlic, ;-( I think you are probably right that fresh garlic would be more potent than the less tasty, dried ground up stuff sold in bottles. Nobody, however, should ever delude themselves into thinking that by eating some garlic or consuming aspirin each day they can place less emphasis upon reducing their fat consumption, neglect increasing their consumption of dietary fiber, vegetables and fruits and pay less heed to the importance of exercise. I think people get their priorities on these things quite wrong, so I hope you don't mind my repeating what you already know. Regards, Ted (1) Ayre (1989). "Garlic preparations and processing. CARDIOLOGY IN PRACTICE, 36:766-8. (2) Mansell & Reckless, "Garlic" BRIT. MED.J., 1991, 303(Aug 17):379-80. (3) Kleijnen et al. (1989). "Garlic, onions and cardiovascular risk factors: A review of the evidence from human experiments with emphasis on commercially available preparations. BR. CLIN PHARMACOL, 28:533-44. (4) Makheja (1979), LANCET (April 7) (5) Bordia (1981). Effect of garlic on blood lipids in patients with coronary heart disease. AM. J. CLIN. NUT., 34:2100-03. (6) Adetumbi et al. (1983). Allium sativum (garlic): A natural antibiotic. MED. HYPOTHESE, 12(3): 227-37. (7) Mattson, F. "Effect of dietary cholesterol on serum cholesterol in man. AM. J. CLIN. NUTR., 35:741. (8) Ernst et al. (1985). "Garlic and blood lipids. BRIT. MED. J. RIT., 291:139. (9). Makheja, A. (1979). LANCET, april 7 (10). Backon, J. (1986). "Ginger: Inhibition of thromboxane synthetase and stimulation of prostacyclin: Relevance for medicine and psychiatry. Med. Hypotheses, 20:271 (11) Elwood et al. "Effect of high-chromium brewer's yeast on human serum lipids. J. AM. Coll. Nutr., 1:263. (12) Anderson & Chen. (1982). "Effects of legumes and their soluble fibers on cholesterol-rich lipoproteins. AM. CHEM. SOC. ABSTRACTS., afgd #39 (13) Malinow et al. (1979). "Alfalfa". Am. J. Clin. Nutr., p. 1810. (14) Mattson et al. (1982) "Optimizing the effect of plant sterols on cholesterol absorption in man. AM. J. CLIN. NUTR., 35:697. (15) Childs et al. (1981). "The contrasting effects of a dietary soya-lecithin jproduct and corn oil on lipoprotein limids in normolipidemic and familial hypercholesterolemic subjects. ATHEROSCLEROSIS, 38:217. (16) Taussig & Heper (1979). Bromelain: Its use in prevention and treatment of cardiovascular disease: Present status. J. INT. ACAD. PREV.MED., 6(1). SERUM FERRITIN AND VEGETARIANS Ted Altar TOPICS BRIEFLY COVERED: 1. Introduction 2. The Remarkabe Molecule of Ferritin 3. Transferrin: Something Not to Confuse with Ferritin 4. Trhansferrin Saturation 5. Serum Ferritin 6. Stages of Iron Deficiency 7. Vegetarians And Iron Deficiency: Some Data 8. Data on Vegetarian Serum Ferritin Levels 9. Vegetarians And Serum Ferritin Levels: Summary 10. Iron And Resistance To Infection 11. Summing Up: Conclusions; Some Practical Advice; References. 1. INTRODUCTION Discussion here on iron might be getting overly technical, so permit to try to provide some background information on some of terms under discussion both for others and myself. I will provide some data that pertaining to the vegetarian's ferritin levels. In turn, questions will be raised about how we are to best answer this worry about whether or not vegetarians have lower iron stores and whether such lower iron stores would make for an increased risk of infection for vegetarians. 2. THE REMARKABLE MOLECULE OF FERRITIN Nature has provided animals with a remarkable protein, called FERRITIN, that proficiently stores iron not needed for daily metabolism. A single ferritin molecule is comprised of 24 smaller protein molecules, bound together forming a hollow ball ideally suited for storing up to 2000-4000 atoms of iron in the center of this ball. Soluble ferrous iron enters the ball through "tunnels" in the outer protein shell. Once the ferrous iron enters into the interior it is then oxidized to form ferric iron which is now insoluble (at least at the pH level of the cell) and in this form it can be densely deposited inside our ferritin storage ball. Odd to think that we have little protein balls filled with iron in our tissue. It this what Bismark meant by "blood and iron"? :-) The purpose of ferritin is to serve as an important storehouse of iron, and one that allows animals to survive for long periods without dietary iron. Storage levels would be determined by the amount of ferritin and the how much iron it contains. Ferritin is therefore an important "store" of *available* iron. Up to 2/3 of the body's stores are in the form of ferritin, the rest is hemosiderin (a form of ferritin but more variable in structure). About 25-30% of the body's total iron for males is to be found in the iron stores. Less of course for woman due to their greater physiological requirement of iron before menopause. It turns out that these stores do vary widely between normal adults (100mg to over 1000mg of iron), males having larger stores (600-1200mg) than women 100-500mg). The bulk of this ferritin iron storage is to be found in the liver, bone marrow and spleen. Relevant to the issue about how we are to determine iron status with respect to available stores, there are different methods that have been used to estimate a person's bodily store of iron, each with their own respective advantages and disadvantages. Here are few: 1. According to Cook (1974), the most precise quantitative method is to draw off blood at weekly intervals until anemia develops. The amount of iron in the blood that necessary to draw off to produce the anemia gives an measure of how much iron was originally stored. You can see the kinds of practical problems such an method would entail; whose is going to volunteer to be have their blood repeatedly bled until they become anemic? Count me out! 2. Another direct method is to take a sample of bone marrow or liver, stain it for iron and examine it under the microscope for iron content. Again, there is a practical problem of obtaining such biopsies. Ouch! 3. A third procedure is to measure iron absorption which varies inversely with iron stores. Radioiron absorption "are currently regarded as the most sensitive indirect measure of iron stores" (Cook et al., 1974). Of course, it is difficult to convince large numbers of people to consume radioisotopes. 4. A fourth method, to be discussed in the next section, is that of serum ferritin levels. While not as precise or always as reliable as some of the above methods, it is much easier to get people to donate a small sample of blood. 3. TRANSFERRIN: SOMETHING NOT TO CONFUSE WITH FERRITIN Lest there be some misunderstanding between similar sounding terms, keep in mind that most (if not all) of the iron that actually circulates in the plasma is bound to something called *TRANSFERRIN*. Not to be confused with ferritin, transferrin is another protein but it is only able to bind at most 2 atoms of ferric iron. It's purpose is not to store iron but to transport it and deliver it to those cells needing iron. These are the local delivery boys for iron. Talk about total SERUM IRON levels usually refers to the amount of plasma transferrin. Normal Levels Of Plasma Iron For Adults males females Average 125 ug/dl 110 ug/dl Range 80-150 70-130 [from Bernat, '83, p.72] In order to understand some of the data on vegetarians that will be presented, we need to need in keep in mind some of these measures of iron status like that of the total plasma iron levels here being defined. Some factors that will lower these plasma iron levels: - Time of day: highest in morning and lowest (57% drop) in evening - Stress: Can be chemically induced (e.g., cortisone, histamine). - Toxins, vaccines and protein decomposition products - Lower hemoglobin production and degradation 4. TRANSFERRIN SATURATION Besides how much total transferrin (plasma iron) there is, another measure of iron status is in terms of how much iron the plasma transferrin molecules are actually carrying. This is referred to as the % of "TRANSFERRIN SATURATION". If all the transferrin molecules only contained 1 atom of iron, then we would speak of a 50% transferrin saturation. The values in healthy adults actually fluctuate quite markedly from a high 47% in the morning to about 13% at night. Researchers would therefore only draw blood for transferrin saturation measures in the morning or early afternoon when low values could be better interpreted as representing a low iron status. Values below 16% are generally considered expressive of iron deficiency. 5. SERUM FERRITIN If most of the stored iron (not the circulating iron) is in the liver, bone marrow and spleen, then what is the point of talking about ferritin in the serum? It turns out that minutes amount of ferritin is also found in the serum and levels of this *serum* ferritin are highly correlated with total iron stores of tissue ferritin and hemosiderin. The correlation, however, is NOT perfect. For example, the correlation of SERUM ferritin levels and the method of iron absorption measurements (mentioned above) is r = -58. Not perfect by far. Part of the problem is due the imperfect test- retest reliability of these measures (.76 for repeated iron absorption measures and .88 for repeated serum ferritin measures). Ok, there is going to be some slop in our measurements of SERUM ferritin but it still is a useful index of total iron stores. Mean Serum Ferritin Levels For Adults males females 90-95 ug/l 25-30 ug/l [from Cook & Skiken, 1982] The range of serum ferritin in normal adults is between 15 and 300 ug/l. A SERUM ferritin level below 12 ug/l is considered diagnostic of iron deficiency (Cook & Skiken, 1982). Others have put the critical value to be below 10 or 12 ug/l (Dallman, 1982). Others might put it at <15 ug/l. It all depends on the importance placed upon identifying people who really are suffering from a low iron status vs the cost of mis-identifying people whose iron status at such levels is normal for them. It should be remembered that high levels are not necessarily indicative of an absence of iron deficiency since "levels as high as 60ug/l may be seen when iron deficiency and inflammation or liver disease occur in the same patient" (Cook & Skiken, 1982, p. 1182). Infections generally lower serum ferritin levels, but they may not be below the 12 ug/l cutoff. This could be important when we come to assess any data about people with repeated infections. Were the measures of serum ferritin taken long enough after the last infection, and if not, then they could well be iron deficient even though their levels are within the normal, or low end of the normal range. Such low end values would not be predictive of a greater susceptibility to infection, but rather, could be the fact of having a recent or concurrent infection that is making for the low levels, but levels not below 12 ug/l. 6. STAGES OF IRON DEFICIENCY There different stages of iron deficiency and there are different ways to differentiate these stages. I have already listed some in my earlier postings on iron. The follow differentiation (Scimshaw, 1991), is as good as any and will serve our purpose: stage 1: Stored iron is depleted, a process reflected in declining levels of ferritin. stage 2: Next, levels of serum iron plummet, and as a result the iron transport protein, transferrin, is no longer fully saturated. At this second stage, cellular compounds requiring iron begin to be affected. stage 3: As the deficiency persists, synthesis of hemoglobin is inhibited, and anemia develops. This last stage is characterized by reduced numbers of now small, pale blood cells. It is generally thought that haemoglobin levels (amount of a iron-bound oxygen carrier protein in the blood) is one way to indicate iron status and the likelihood of anemia being present (Victor Herbert, 1987).: haemoglobin men < 13 g/l women < 12 g/l The BMA (Brit. Medical Association), however, stated in 1986 that haemoglobin levels as low as 11 g/l are NOT significant to health and that the importance of maintaining haemoglobin at the previously recommended high levels has been exaggerated. [see DIET, NUTRITION & HEALTH, London: BMA, 1986.] T TESTS FOR IRON DEFICIENCY Serum ferritin: Normal --> 20-250 ug./L. (ug./L. = ng./dL.) Low --> 12-19 ug./L. Iron-deficiency --> fewer than 12 ug./L. Iron-deficiency anemia --> fewer than 12 ug./L. AND hemoglobin levels below normal Transferrin saturation: Normal --> 16-55 (as a percentage of total iron binding capacity of serum) Iron-deficiency --> less than 16 Iron-deficiency anemia --> less than 16 AND hemoglobin levels below normal TESTS FOR ANEMIA (but not specifically iron-deficiency anemia) Hemoglobin: Normal Men --> 13.5-18 g./dL. Women --> 12.5-17 g./dL. Anemia --> levels below normal Hematocrit: Normal (as a percentage of red blood cells in the total blood volume) Men --> 40-54 Women --> 37-47 Anemia --> levels below normal 7. VEGETARIANS AND IRON DEFICIENCY: SOME DATA Now, there is no evidence that vegetarians or vegans do suffer from an higher incidence of iron anemia (stage 3) than omnivores. This is not to say the iron anemia is not a serious problem but rather that it is a problem for BOTH omnivores and vegetarians, and vegetarians are no worst off. I cited in my earlier postings on iron the U.S. incidence rates from the NHANEII study in order to quantify how much of a problem or worry this really is. With respect to stage 2, studies of vegetarians do show that their levels of transferrin saturation to be about the same as omnivores. Here is some data to support this conclusion. STUDY #1 Long-term SDA Vegetarian Women (mean age 52.9) with without Normal supplementation supplementation values # of subjects 11 40 transferrin saturation 39% 36% 37% hemoglobin (g/dl) 12.9 13.2 14+/-2 plasma iron (ug/dl) 135 107 50-150 TIBC (ug/dl) 346 312 250-410 [from Anderson et al.,1981] STUDY #2 Iron Status Of Males, Aged 21-52 Meat-eaters Vegetarians # of subjects 18 36 dietary iron (mg) 18 17 Plasma iron (ug/dl) 104 116 Iron binding capacity (ug/dl) 373 393 transferrin saturation (%) 28 +/-5 30 +/-7 [Latta & Liebman, '84] According to the Nutrition Canada Interpretive Standards, serum transferrin values < 20% would indicate a risk for iron deficiency. Note that vegetarians are not iron deficient according to either this index of serum transferrin saturation or total plasma iron levels. More data below will further confirm this observation. In my previous postings on iron, we saw that plasma iron and hemoglobin levels of vegans were in the normal range. The above two studies (and the studies mentioned below) supports that the same is occurring with lacto-vegetarians. With respect to stage 3 and stage 2 of iron deficiency, the evidence here presented would support conclusion that the vegans and vegetarians are no different than omnivores on these measures of iron status. 8. DATA ON VEGETARIAN SERUM FERRITIN LEVELS The debate now boils down to stage 1. Low levels of serum ferritin are correlated with low iron stores. Do vegetarians fare worst than omnivores, and if so by how much and is this a difference that makes a difference in terms of one's health being compromised? These are the tough questions that can be pose. Ok, let's first look at the data on vegetarians. Now, I don't on hand any data regarding the ferritin levels for vegans, but it should be noted that vegans do consume more iron and vitamin C than lacto-ovovegetarians: Intakes as a Percentage of RDA (male and females, ages 27 - 49 yr) Iron (% RDA) Vitamin C (%RDA) vegan (10) 186.3 704 lacto-ovovegetarian (9) 158.4 639 whole food omnivore (8) 125.3 485 average omnivore (10) 128.2 351 [from Carlson et al, 1985) It is true that along with consuming more iron and vitamin C, vegans also eat more dietary fiber than lacto-ovovegetarians which can interfere with absorption. The role of fiber on iron absorption is conflicting. One reason might be that low levels of fiber infer with absorption while high levels don't. In one experimental study with volunteers (Kelsay et al., 1979) it was found that when given diets with the same amount of iron, the iron status of the volunteers was actually MORE favourable than when the fiber intake was 59g/day (vegans consume about 45g/day) than when it was only 9g/day. Also, vegans don't have the disadvantage of egg yolk and cows milk that also hinders iron absorption. It is likely, therefore, that vegans have a better iron store status than lacto- ovovegetarians, but I don't have on hand any data on the serum ferritin levels of vegans. Now let's look at the results for lacto-ovovegetarians. Here we have more data which is to be expected since there more lacto- ovovegetarians than vegans. Unless otherwise indicated, the data below pertain to vegetarians not taking any supplements of iron. STUDY #3 (female college students) omnivores vegetarians subjects 9 13 analyzed iron intake (mg/d) 11.64 13.86 ascorbic acid intake (mg/d) 80 131 hemogoblin (g/dl) 14.24 13.92 hematocrit (%) 42.22 41.58 serum iron (ug/dl) 140.00 120.62 serum ferritin (ng/ml) 41.1 22.58 [McEndree et al., 1983] STUDY #4 (males & females, mean age 29.3 yr) omnivores vegetarian subjects 37 93 ferritin (ng/ml) 70 (sd=61) 45 (sd =33) males 58 (sd = 34) females 30 (sd = 22) [Helman & Darton-Hill, '87, 45: 785] This involved a study of *new vegetarians* as defined by a population of people who had adopted the diet relatively recently and often in association with a generally alternative life style philosophy. How many belonged to a Macrobiotic sect was not reported. None were vegans. This study of new vegetarians who have been practicing their diet for a few years (mean 7.3 years) did show that a higher percentage of women vegetarians (27%) ) as compared to woman omnivores (12%) were at low ferritin levels defined as levels below <15 ng/ml. A slightly lower percentage (5%) of male vegetarians were low by this criterion as compared to male omnivores (8%). STUDY #5 Premenopausal women meat eaters vegetarians fish/poultry number of subjects 16 15 20 mean age 31 yr 27.9 28.9 mean mensural d/yr 58.4 57.0 61.2 dietary iron (mg) 12.7 11.8 11.9 ascorbic acid (mg) 144 169.3 168.6 serum ferritin (ng/ml) 29 21 14 [Worthingon-Roberts et al., '88] ** Note that the fish/poultry eaters had the lowest serum ferritin levels, not the vegetarians. STUDY #6 And we might just as well look at the Kies & McEndree (1982) study I cited in my earlier postings on iron: Consider the following table of the iron nutritional status status of one sample of Seventh Day Adventists students. Iron Nutritional Status of Omnivore and Vegetarian Students Eating a Lacto-Ovo-Vegetarian Food Service Diet (Self- Selected). omnivores vegetarians # of Participants 13 15 Serum iron (ug/dl) 121 124 Serum ferritin (mg/ml) 36.48 24.84 Hemoglobin (g/dl) 14.05 13.95 hematocrit (%) 42.04 41.67 Iron intake (mg/day) 10.36 10.60 Ascorbic acid (mg/day) 95.6 125 [from Kies & McEndree,1982] STUDY #7 This next study I will not include in my summary table below because we are here dealing with an anomalous group of East Indian "semi-vegetarian" woman (majority ate meat, poultry & fish no more than twice a week). Semi-vegetarian East Indian Immigrants to Canada Serum Ferritin females 18-39 14.8 females 40-64 18.3 males 18-39 70 males 40-64 91.4 [Bindra & Gibson, '86] This result may be due to the type of diet in which fruits and vegetables don't figure as prominently as with other Canadians (vitamin C intake was 121mg for the women and 135 for the men). Also, the high phytic acid and fiber content of their chapatti breads, the tannins in their tea, tamarind, chillies, turmeric, and coriander, plus the high intakes of calcium from milk were thought by the authors to be counteracting the body's ability to absorb more iron according to physiological need. Certain problems in adjusting to the dietary habits and food supplies of a new culture should also be considered as posing a particular difficulty for these women that might have resulted in their problematic diet. We should keep in mind that there is a diversity of vegetarian diets and certain groups of "vegetarians" should not presume that the favourable nutritional status of other vegetarian groups will apply to them. In particular, macrobiotic "vegetarians", or this group of new East Indian Canadian "semi-vegetarians" should be considered separately. 9. VEGETARIANS AND SERUM FERRITIN LEVELS: SUMMARY To briefly summarize the data on serum ferritin levels, we have: Summary Of Studies Recording Serum Ferritin Levels meat eaters (ng/ml) vegetarians (ng/ml) Worthingon-Roberts et al.,'88 29 21 McEndree et al., '83 41 23 Kies & McEndree, '82 36 25 Helman & Darton-Hill, '87 70 45 All levels are within the normal range. Serum ferritin levels for vegetarians are lower than omnivores, suggesting that vegetarian students might have lower iron stores. Serum ferritin, while highly correlated with iron stores, is not a perfect measure of iron stores but rather an index of those iron stores. The average of serum ferritin is still within the normal range, that is, it does not fall for the majority of lacto- ovovegetarians below the cutoff of < 12ng/ml associated with iron deficiency or even the higher cutoff of < 15 ng/ml that the is sometimes used. Other blood iron indices were no different from that of omnivores. Serum ferritin levels are lower level for vegetarians than for omnivores. The average levels, however, are not below the usual cutoffs for low serum ferritin levels. A small proportion of both omnivores and lacto-ovovegetarians do fall below a cutoff of < 15 and there is a greater percentage of vegetarians than omnivores that do fall below this cutoff. Interesting, the lacto-ovovegetarians are in some studies consuming about the same amounts of dietary vitamin C as do omnivores. This suggests that lacto-ovovegetarians are not eating as well as vegans and do have this greater of risk incurring low iron stores. The next question is whether this difference is a difference that makes a difference. We are not talking about iron anemia, but about a sub-clinical iron deficiency which may concievably compromise one's ability to resist and fight off infections. What is the evidence for this worry? 10. IRON AND RESISTANCE TO INFECTION We know that serum ferritin levels are lower in vegetarians than in omnivores (about 1/3 lower). Is this good or bad? On the "con" side we have the worry that an elevated iron status may increases the risk for CHD (1992 Finish study) and cancer (Selby & Friedman, '88; Weinberg '81; Stevens et al., '88). As one team of researchers concluded: "Too little iron is clearly detrimental. However, iron elevated beyond a level necessary to avoid anemia may also have adverse consequences. Our studies have attempted to focus on the effect of variations in iron stored on the risk of cancer. These variations may be well within the `normal' range in otherwise healthy persons. If elevated iron stores increase the long-term risk of cancer, and if iron intake affects iron stores, then the policy of iron fortification of food should be reconsidered. Iron supplementation for those who are not anemic may be unwise". [from Stevens et al., '88] On the "pro" side, there are some interesting studies that indicate a greater incidence of certain bacterial infections being correlated with sub-clinical iron deficiency. In these studies, however, the measure of iron deficiency was NOT based upon a measure of serum ferritin levels. Before we look at a few results, note that Dallman (1987) in his review noted that one of the key problems with the research showing an association between iron deficiency and the frequency of infections was that such associations were sought: . . . usually among poor populations. Both iron deficiency and infections are known to be associated with poverty and lack of education (and probably also with crowded housing and contaminated water supplies). Consequently if an iron-deficient population is found to have a high rate of infection, it is difficult to prove that iron-deficiency played an independent role from that of poverty, poor education, and associated adverse socioeconomic factors. Dallman is saying that while the effects of iron deficiency (beyond stage 1) on work performance or cell-mediated immunity are well established, the effects on the incidence of, and population susceptibility to, infection are less clear, simply because of the difficulty of carrying out properly controlled studies. STUDY A There is a study published in Lancet (Weijmer et al, 1990), it which there a observed relationship between low serum iron concentrations that was associated with the increase susceptibility to the bacterial infections observed. The authors concluded: "The role of iron in susceptibility to infection remains unclear, but our findings in patients with recurrent staphylococcal furunculosis indicate that there may be a relation between low serum iron concentrations and susceptibility to infection." Two points to note about this study. First, the selection of patients with the recurrent bacterial infection (staphylococcal furunculosis) seems to have occurred at the start of this study (the write up is not very clear on this), and then the patients were followed over time. If this is the case, then what we have is really a case control study and we can't be sure if the recurrent infections themselves didn't make for low plasma iron levels rather than the other way around. A second point to keep in mind is that the data on vegans and lacto-ovovegetarians indicates them to have comparable plasma iron levels. The patients in this study had plasma iron levels that were very low indeed, around 54 ug/dl (remember, the normal range was 70-130 for females, 80-150 for males). STUDY B. What about the study (Bondestam et al., '85) on some young children (mean age 3.5 yr) that suffered from frequent infections of the respiratory tract and middle ear? Most of these children who were highly susceptible to infection had normal levels of plasma iron (18% had levels below normal), but their normal levels were lower than those of the control group of children who did not have an undue susceptibility to infection. Recurrent Infections Plasma Iron (ug/dl) upper respiratory tract 114 lower " " 110 middle ear 106 controls 140 Does this mean that low iron stores within the normal range makes for an increased susceptibility to infection. Maybe, but all we have is a correlational study, one dealing with young children whose average plasma levels are lower than adults: "It is postulated that the changes in the trace element status observed in this study are the combined effects of infection and malnutrition . . . the possible role of trace element deficiency in predisposing to or perpetuating undo susceptibility infection in children remains to be elucidated." The actual role of low plasma iron does not seem to be something that the authors are themselves affirming with much certainty. Also, we simply don't know what kind of malnutrition is here implicated, if any; it could be a protein-energy malnutrition or a zinc deficiency (here too zinc levels are also induced by infections) or a combination thereof. OTHER STUDIES. Other studies, however, show that too much iron in the body increases the risk of infection (Emery, 1991) How much of a deficiency as indexed by serum ferritin is required to make for an increase susceptibility to infection is maybe the next question to ponder. Some idea would be obtained from the studies showing a relationship of a low iron status with an increased infection rate. However, as Reviewer of the research has pointed out: Within the last 2 decades, hundreds of articles have appeared in the medical and biochemical literature that strongly supports the hypothesis that too much iron in the body increases the risk of infection . . .There is no doubt that severe iron deficiency also weakens the body to such an extent that many clinical problems, including infection, are greatly enhanced. . . . Nevertheless, there is often considerable difficulty in interpreting the data to allow one to state definitely that the cause of the disease is attributable uniquely to iron deficiency. In most populations of the world, inadequate dietary iron goes hand in hand with other nutritional deficiencies. Iron is required for the synthesis of proteins, but so are dietary amino acids. Consequently, it is almost predictable that in a large fraction of the undernourish population of the world, we will find iron deficiency, but that does not prove that clinical manifestations are in any way directly due to lack of adequate iron. [Thomas Emery, 1991] Another consideration that here bears on the this issue: There is now general agreement that the combination of anemia, low serum ferritin, and an additional laboratory abnormality (mean corpuscular volume, transferring saturation, or protoporphyrin) constitutes strong evidence of iron deficiency. Anemia and low transferrin saturation alone are ambiguous since they may also be present not only during severe and mild infections but also for SOME TIME AFTER the clinical manifestations of an infection have been resolved. [emphasis mine] [Dallman, 1987] If diagnosing an iron deficiency that results in anemia is difficult, would we not expect a similar and maybe greater difficulty when it comes to diagnosing a sub-clinical low iron status? Serum ferritin, while a convenient measure, is not the best measure of extant iron stores. Finally, and more important, how low is too low? Dallman would state: A low serum ferritin by itself indicates low iron reserves and would not be expected to be associated with any physiological handicaps. [Dallman,1987, p. 332] What then is the optimal level of iron stores? Is more always better? Dallman makes an interesting comment about how we must conceive of what would be an "optimal" iron status: The possibility that iron deficiency could both impair and enhance the immune response has been described as presenting a paradox or dilemma. We are accustomed to the idea that the deficiency of a nutrient is harmful in all respects. However, in the case of iron (and probably other nutrients such as unsaturated fatty acids), it is becoming difficult to avoid the conclusion that the optimal range of intake differs according to the conditions under which a specific body or cell function is examined. For example, in a protected environment an iron intake low enough to impair oxygen delivery and ATP production may actually prove to decrease mortality and morbidity with certain types of infections. It is doubtful that the same would be true under most REAL LIFE circumstances that require the body to respond to very many different environmental stimuli and hazards. For this reason, what we consider the optimum intake for a nutrient increasingly can be seen to represent a composite or compromise among often conflicting values for individual body functions and environmental conditions. [p. 333] 11. SUMMING UP: CONCLUSIONS; SOME PRACTICAL ADVICE; REFERENCES. So what can we conclude about vegetarians and an increased susceptibility to infection? 1. Vegetarians do have lower serum ferritin levels than omniovres, but the average level is still within what is considered the normal range for healthy persons. 2. A lower plasma iron is, however, associated with an increased susceptibility to infection, but vegetarians don't have seem to have a consistently or markedly lower serum iron level to that of omniovres. 3. We don't have evidence that subclinical lower serum ferritin levels is also associated or causative of an increased infection susceptibility, nor how low such levels would have to be (if there is a threshold) for such an increased susceptibility to occur. 4. We don't any direct evidence that vegetarians do in fact suffer from a higher incidence of bacterial infections or a compromised ability to fight off those infections. The notion that vegetarians might be at a higher risk for infection is simply not supported. Just to set the record straight: I am not here suggesting, nor have I ever suggested, that vegetarians (or omniovres) should neglect paying attention to eating quality foods with a high iron and vitamin C content. Quite the contrary. Lacto-ovovegetarians are not eating as well as vegans in this respect and should by all means pay attention to this fact. Generally, almost everybody's diet contains enough iron, but the problem has always been that only a fraction of that iron is absorbed. The interesting conclusion here is that iron "deficiency is almost always caused by a low bioavailability rather than by a low intake" (van Dokkum, 1992). The Dutch Nutrition Council has estimated an iron absorption for the average omnivorous Dutch diet to be 12% while that of a vegetarian diet has been set at 8% (ibid). Of course, we should here be cautious about relying on such single numbers to define overall dietary iron absorptions for all individual or sup- populations. Many factors come into play to determine absorption, and the absorption of non-heme iron can range from 1% to 40% (Health & Welfare Canada, 1990). As we know, the concurrent ingestion of such things as vitamin C and fructose can enhance absorption while such food items as tea, coffee, egg yolk and wheat bran can inhibit absorption. One of the key individual factors to keep in mind is that absorption increases with decreasing stores. It was shown in Swedish men that absorption increased from 21% to 74% as bone marrow hemosiderin decreased (Magnusson et al., 1981). Even when serum ferritin values were somewhat low (24 ug/l) absorption was 49% as compared to a lower 21% absorption when serum ferritin values were at the high end of 79 ug/l (ibid). Having warned you about using these single values that would estimate the overall average dietary absorption of iron, permit me to ignore my own warning and play with this single figure of 8% absorption for vegetarians. The US RDA's for iron assumes an average daily absorption of 10% for all individuals. If we now provisionally accept this 8% figure as the average daily absorption of dietary iron by vegetarians, then a vegetarian should therefore aim to consume 1.25% of US RDA values for iron. As I documented in my earlier postings, vegans are already consuming about twice the RDA's for iron and are consuming about 45% more vitamin C than omnivores. The situation for lacto- ovovegetarians females is, however, not as favourable. The following table summarizes the dietary intakes for lacto- ovovegetarians from those studies that provide data for both iron and vit C daily intakes: Summary of Lacto-ovovegetarians Intakes of Iron and vit C. Study sample sex Iron(mg) Vit C (mg) Carlson et al. '85 (9) M&F 158%Brit RDA 639% RDA Pedersen et al. '91 (34) F 20 316 Nieman et al '89 (23) F 12 155 Worthington-Roberts '88 (20) F 12 169.3 McEndree '83 (13) F 14 131 Taber & Cook '76 (13) F 18 194 " " (15) M 21 217 Hardinge & Stare '54 (15) F 16 220 " " (15) M 22 250 ==== ===== weighted average (excluding Carlson) 16 mg 273 mg We can see that the men, whose RDA for iron is only 10mg, are easily meeting this need and more. Given a 8% absorption we would expect that men would need a RDA of 12.5mg, which again is being easily met and exceeded. From these studies, an overall weighted average (weighted for sample size) of dietary iron intakes for lacto-ovovegetarian women is 15.7mg. This meets the American RDA of 15mg for woman between the ages of 19 - 51 yrs (Canada has set it's RDA to 13mg/day). At a 10% absorption rate this should be fine, but given the Dutch estimate of an 8% absorption, then our vegetarian woman should be consuming about 18.75mg of iron, which most are not. We are about 3mg short (which is about the amount found in a cup of chard or a cup of peas or 2 cups of strawberries). Does this mean that the average lacto-ovovegetarian woman is going to inevitably suffer progressively first from low stores and then eventually iron anemia? Should we be observing a lot of vegetarian woman slowly shuffing around with deformed finger nails? Apparently not, given our data on hemoglobin and other blood indices of iron being normal for vegetarians. So what gives? Well, presumably the body is not stupid and we know that iron absorption increases with need. The intestinal mucosa will actually change to meet this need so as to be able to more readily absorb iron. Does this mean that vegetarian woman need not worry? Of course not. Both omnivore and vegetarian woman must take this worry seriously and examine their dietary and life-style habits, monitor their health and consider a blood test. To this end, I have repeated tried to bring to everyone's attention two important considerations. Lacto-ovovegetarians should be consuming more iron rich foods, as well as, more vitamin C rich foods. Absorption will increase with need and with certain dietary enhancers. I've listed in my previous notes on iron the main factors that either enhance or inhibit iron absorption and these should be given careful attention. Vitamin C is the most important enhancer but it must be present at the time in each meal from which one hopes to obtain iron. Forget about the Diamond's nonsense about food combining which would have us forgo fruit for dinner or lunch or the Macrobiotic nonsense that would have us eliminate, or near eliminate, all fruits altogether. Obviously, do eat more fruit and vegetables, which contain both iron and vitamin C. If one needs to take a vitamin C tablet at each meal, then so be it. Apparently a mere 75 mg of ascorbic acid has a maximal effect on absorption and amounts in excess of 75 mg apparently has no further benefit for increasing the absorption of iron (Monsen, 1982; Hunt & Groff, 1990). This amount of 75mg of vitamin C is easily obtained in a single cup of cooked kale, or a cup of strawberries, or a cup of unsweetened orange juice. REFERENCES: Anderson et al. (1981). The iron and zinc status of long-term vegetarian women. AMER. J. CLIN. NUTRITION, 34:1042. Bernat (1983). IRON METABOLISM. N.Y.: Plenum Pr. Bindra & Gibson (1986). Iron status of predominantly lacto-ovo vegetarian East Indian immigrants to Canada: a model approach. AMER. J. CLIN. NUTR., 44:643. Bondestam et al. (1985). Subclinical trace element deficiency in children with undue susceptibility to infections. ACTA PAEDIATR. SCAND., 74:515. Carlson et al. (1985). A comparative evaluation of vegan, vegetarian, and omnivore diets. JOURNAL OF PLANT FOODS, 6:89 Cook et al (1975). Serum ferritin as a measure of iron stores in normal subjects. AMER. J. OF CLIN. NUTR., 74:681. Cook & Skikne (1982). Serum ferritin: A possible model for the assessment of nutrient stores. AM. J. CLIN. NUTR., 35:1180 Dallman (1982). Biochemical and hematologic indices of iron deficiency. In Pollitt & Leibel (ed), IRON DEFICIENCY: BRAIN AND BEHAVIOUR. N.Y.: Raven Pr. Dallman (1987). Iron deficiency and the immune response. AM. J. CLIN. NUTR., 46:329 Emery (1991). IRON AND YOUR HEALTH. Florida: CRC Pr. Health & Welfare Canada. NUTRITION RECOMMENDATIONS: THE REPORT OF THE SCIENTIFIC REVIEW COMMITTEE, 1990. Helman & Darton-Hill, (1987). Vitamin and iron status in NEW vegetarians. AM. J. CLIN. NUTR., 45: 785] Hunt & Groff (1990). ADVANCED NUTRITION AND HUMAN METABOLISM. New York: West Pub. Co. Kelsay et al. (1979). Effect of fiber fruits and vegetables on metabolic responses of human subjects. AM. J. CLIN. NUTR., 32:1876 Kies & McEndree (1982). Vegetarianism and the bioavailability of iron. In Kies (ed.), NUTRITIONAL BIOAVAILABILITY OF IRON. Washington: Amer. Chemical Soc., 1982. Latta & Liebman, (1984). Iron and zinc status of vegetarian and nonvegetarian males. NURT. REP. INTERN. 30(1):141] Levin et al. (1986). Mineral intake and blood levels in vegetarians. ISRAEL J. OF MEDICAL SCIENCES, 22:105. Magnusson et al. (1981). Iron absorption in relation to iron status. SCAND. J. HAEMATOL., 27:201 McEndree et al.(1983). Iron intake and iron nutritional status of lacto-ovo-vegetarians and omniovre students eating in a lacto-ovo-vegetarian food service. NUTR. REP. INTERN., 27(1):199] Monsen (1982). Calculating dietary iron bioavailability: refinement and computerization. J. Am. Diet. Assoc., 80:307. Nieman et al. (1989). Dietary status of Seventh-Day Adventist vegetarian and non-vegetarian elderly woman. J. AMER. DIETETIC ASSOC., 89:1763. Pederson et al. (1991). Menstrual differences due to vegetarian and nonvegetarian diets. AM. J. CLIN. NUTR., 53:879 Scimshaw (1991). Iron Deficiency. SCIENTIFIC AMERICAN, Oct 1991:46 Selby & Friedman (1988). Epidemiologic evidence of an association between body iron stores and risk of cancer. INT. J. CANCER, 41:677. Stevens et al. (1988) Body iron stores and the risk of cancer. N. ENG. J. MED., 319:1047 Taber & Cook (1980). Dietary and anthropometric assessment of adult omnivores, fish-eaters and lacto-ovo-vegetarians. J. AMER. DIET. ASSOC., 76:21 van Dokkum (1992). Significance of iron bioavailability for iron recommendations. BIOL. TRACE ELEMENT RES., 35:1 Weijmer et al. (1990). Preliminary report: furunculosis and hypoferraemia. LANCET, 336 (Aug, 25):465. Weinberg (1981). Iron and neoplasia. BIOL TRACE ELM. RES., 3:55. Worthingon-Roberts et al. (1988). Iron status of premenopausal woman in a university community and its relationship to habitual dietary sources of protein. AM. J. CLIN. NUTR., 47:275]