Ratio of Low Serum Zinc Levels in Elderly Japanese People Living in the Central Part of Japan

ORIGINAL ARTICLE Ratio of low serum zinc levels in elderly Japanese people living in the central part of Japan

M Kogirima1, R Kurasawa2, S Kubori2, N Sarukura3, M Nakamori3, S Okada4, H Kamioka5 and S Yamamoto3

1Department of Food Science and Nutrition, Doshisha Women’s College, Kyoto, Japan; 2Kita-mimaki Onsen Clinic, Tomi-city, Nagano, Japan; 3Graduate School of Health Biosciences, Tokushima University, Tokushima, Japan; 4Laboratory of Physical Education and Medicine, Tomi-city, Nagano, Japan and 5Department of Regional Environmental Science, Tokyo University of Agriculture, Tokyo, Japan

Objective: Elderly patients often have impaired taste, nausea, anorexia and delayed healing of decubitus. In many of these patients, serum zinc levels are low and they respond to zinc supplementation. To date, no epidemics of zinc deficiency have been reported in Japan. We studied the prevalence of zinc deficiency and its causative factors in a typical local town in Japan. Design: Cross-sectional study. Setting: Nagano Prefecture (central Japan). Subjects and methods: Serum zinc levels were measured in 1009 habitants (18–96 years old) who participated in an annual mass health examination program of the City in 2003. Of all subjects, 86 with low serum zinc (o65 mg/dl) or high zinc (X90 mg/ dl) levels were randomly selected, and a dietary survey using 24-h recalling methods for 2 consecutive days was performed in 2004. Among them, blood was collected in the morning from 50 subjects (26–94 years old). Results: The percentage of those with low serum zinc levels (2.5th percentile of Americans) was 37.9% in the elderly (X60 years old). The age-adjusted prevalence of low serum zinc was 21.1% in the aged population. A positive correlation was detected between zinc intake and serum zinc levels in elderly subjects (X60 years old). Conclusions: The possibility of zinc deficiency in adult inhabitants in central Japan rises with age. The deficiency correlates with dietary zinc intake. Sponsorship: This research was supported by a grant from local government officials in Tomi City (2004) and the funding of Fancl Corporation. European Journal of Clinical Nutrition (2007) 61, 375–381. doi:10.1038/sj.ejcn.1602520; published online 13 September 2006

Keywords: serum zinc; zinc deficiency; zinc intake; elderly population; aging

Introduction pregnant women, infants and children (Bates et al., 1993; Goldenberg et al., 1995; Brown et al., 1998; Bhutta et al., There is a significant amount of evidence showing that zinc 1999; Caulfield et al., 1999; Shankar et al., 2000; Osendarp deficiency effects health (Aggett, 1989; Prasad, 1991; Gibson et al., 2001; Black et al., 2004). Zinc deficiency has also been et al., 1998; Hambidge et al., 2000; Osendarp et al., 2001; reported in industrialized countries. The Third National Hopley et al., 2004; Lee et al., 2004). Zinc deficiency is a Health and Nutrition Examination Survey (NHANES III) of serious problem in developing countries. Many studies have the US showed that children of 1–3 years old, adolescent shown the adverse effects on growth and morbidity as well as females of 12–19 years old and elderly people of more than the prevention of infection by zinc supplementation in 70 years old were at the greatest risk from inadequate zinc intake (Briefel et al., 2000; Ervin and Stephenson, 2002). Correspondence: M Kogirima, Department of Food Science and Nutrition, Another dietary survey in the US, 1994–1996 Continuing Doshisha Women’s College, Imadegawa Teramachi Kamigyo-ku, Kyoto 602- Survey of Food Intakes by Individuals (CSFII), showed an 0893, Japan. association of low zinc intake and age (Ma and Betts, 2000). E-mail: [email protected] Received 9 May 2006; revised 24 July 2006; accepted 30 July 2006; published The Second National Health and Nutrition Examination online 13 September 2006 Survey in the US (NHANES II) showed that serum zinc levels Ratio of low serum zinc levels in elderly Japanese M Kogirima et al 376 were the lowest in young children, then increased steadily blood was collected in the morning from 50 subjects (26–94 with age, peaking between about 18 and 25 years old, then years old). None of the subjects took a zinc supplement. starting to decrease gradually and dropping after reaching Serum zinc levels were measured by a commercial 65–70 years old (Hotz et al., 2003). Serum zinc levels were laboratory (SRL Co., Ltd.) using atomic absorption spectro- significantly higher in males (88.470.2 mg/dl) than in photometry (Hitachi Co., Ltd., Z-6100). The dietary survey females (83.370.2 mg/dl) (Po0.001). was performed according to the 24-h recalling method The latest studies on zinc nutritional status were reported before blood sampling on 2 successive weekdays. The use by ZINCAGE, a specific targeted research project funded by of medicines was also considered. The subjects were asked the European Union (Andriollo-Sanchez et al., 2005; Hill to record ingested foods, beverages and medicines, and a et al., 2005; Hodkinson et al., 2005). There have been few dietitian interviewed each subject to confirm how much was large-scale research projects regarding zinc nutrition in consumed and to identify any foods that had not been elderly Japanese, although the aged population is rapidly reported. The zinc intake was calculated according to the growing. Standard Tables of Food Composition in Japan (The Fifth Approximately 40% of taste impairment was reported to Revised and Enlarged Edition). In addition, 15 questions show low serum zinc levels in Japanese (Yamamoto, 1996). relevant to symptoms of zinc deficiency were investigated. We also reported the relevance of taste impairment and the Questions were appetite, tastes, tiredness, skin condition and low nutritional state of zinc (Takeda et al., 2004). In addition, condition of the stomach, the intestines and the mouth. The some reports indicate that taste impairment in elderly people questions were chosen based on our clinical experience is a problem (Yamamoto, 1996). We often encounter elderly (Kurasawa and Kubori, 2006). patients complaining of taste impairment, stomatitis symp- The purpose and procedures of the study were explained to toms, nausea, anorexia, delayed healing of decubitus, all subjects before obtaining informed consent. behavior abnormalities and others in the daily medical practice in our hospital situated in the same region as this Statistical analysis study (Nagano Prefecture, central Japan). These symptoms Statistical analysis was performed using SPSS software can be improved by the supplementation of zinc (Kurasawa (version 13.0J, SPSS Inc, Chicago, IL, USA). The results were and Kubori, 2006). We decided to examine whether the expressed as mean7s.d. Comparisons between the genders number of people with potential zinc deficiency existed. In or the two groups were performed by unpaired Student’s our 2003 study, we examined serum zinc levels in 1431 t-test. For comparison of the groups according to age, residents (6–96 years old) and found that serum zinc levels analysis of variance was used. Relationships between the decreased with age, and were lower in females than in males variables were analyzed by calculating Spearman’s correla- (Kurasawa et al., 2005). We also observed circadian variation tion coefficients. In all analysis, Po0.05 was considered (high in the morning and low in the afternoon). There are significant. other reports on circadian variation (Hetland and Brubakk, 1973; McMillan and Rowe, 1982; Yokoyama et al., 2000). We examined the prevalence of zinc deficiency and its Results causative factors in Nagano Prefecture, a central part of Japan. Table 1 shows the proportion of the population of these subjects except for the 18- to 19-year-old subjects. We Subjects and methods Table 1 Proportion of the population

Serum zinc levels were measured in 1009 habitants (18–96 Age (years) Subjects Japanese a years old) who participated in an annual mass health examination program of the Kita-mimaki area of Tomi City Male Female Overall in Nagano Prefecture, central Japan, in 2003. Kita-mimaki Nn %% area is a typical rural region in Japan, and the local 20–29 29 48 7.6 16.5 government and the residents cooperated in the study. The 30–39 48 70 11.7 17.7 population of this city was 31 222 of which about 20% were 40–49 41 92 13.2 15.4 over 65 years old. Fifteen percent of the adults in this city are 50–59 48 108 15.5 18.7 engaged in primary industry. 60–69 113 120 23.1 15.3 70–79 82 109 19.0 11.0 Of all the subjects in 2003, 90 with low serum zinc 80– 30 69 9.8 5.5 (o65 mg/dl) or high zinc (X90 mg/dl) levels were randomly selected and agreed to blood sampling again after 1 year. 391 616 100.0 100.0 Although a dietary survey was performed in 2004, four The data from two persons aged under 20 from all subjects were eliminated. subjects refused to participate in the dietary survey. Eighty- aThe author calculated on the basis of estimated Japanese population data in six subjects agreed to all aspects of the survey. Among them, 2003.

European Journal of Clinical Nutrition Ratio of low serum zinc levels in elderly Japanese M Kogirima et al 377 compared the population structure with that of the whole significantly high proportion compared with the other age Japanese population. The proportion of 20- to 29-year olds groups (w2 ¼ 55.4, Po0.001), we calculated the age-adjusted was low and that of those 60 years old or older was high in percentage (Japanese model population structure by age and this study. sex in 1985) and found the incidence to be 21.1%. Table 2 shows the number of subjects in each subgroup for Table 5 shows multiple regression analysis (stepwise the 1009 subjects. There were 392 males (mean age: method) with serum zinc level as the dependent variable 58.1717.1 years, 19–93 years old) and 617 females (mean (n ¼ 1009). The sampling time of blood showed a strong age: 57.4717.8 years, 18–96 years old). The 1009 subjects relation with serum zinc levels. To exclude the confounding were divided into 18 groups by age, gender, fasting status factor of the blood sampling time, we performed a simple and time of blood sampling. correlation analysis only on the subjects with blood Table 3 shows the serum zinc levels of each 18 subgroups. sampling in the morning (n ¼ 443) regarding the relation Regardless of the fasting status, the serum zinc levels were between serum zinc level and age (Figure 2). Although a higher in the morning than in the afternoon (Po0.01). The slight decline was observed, serum zinc levels fell remarkably serum zinc levels in the group of those over 60 years old were with age after 60 years of age (coefficient of correlation in lower than those in the group of 18- to 39-year olds in both genders. Figure 1 shows the percentage of people whose serum zinc 45 Total level was below the lower cutoff point by Hotz et al. (2003) Male shown in Table 4. Among the subjects of 60 years or older, 30 Female 37.9% had low serum zinc levels. As this age group was a 15

Table 2 Number of subjects in subgroups cutoffs (%) low Below 0 18-39 y 40-59 y 60 y≤ Age (years) Serum sampling Overall Male Female Age groups 18–39 Morning fasting 38 19 19 Figure 1 Percentage of subjects with serum zinc concentrations Morning non-fasting 28 8 20 lower than the cutoff point. There were 1009 subjects. Table 2 Afternoon 131 51 80 indicates the details of each number of subgroups.

40–59 Morning fasting 112 46 66 Table 4 Suggested lower cutoffs (2.5th percentile) for the assessment Morning non-fasting 22 6 16 a Afternoon 155 37 118 of serum zinc concentrations in population studies Serum zinc concentration 60p Morning fasting 180 85 95 Morning non-fasting 63 19 44 Males aged Females aged Afternoon 280 121 159 X10 years X10 years 1009 392 617 Morning fasting (mg/dl) n ¼ 3343 74 70 Morning non-fasting (mg/dl) n ¼ 3408 70 66 Afternoon (mg/dl) n ¼ 3268 61 59

Table 3 Serum zinc concentration in subgroups aHotz et al. (2003). Data derived from the second National Health and Nutrition Examination Survey (NHANES II). Total number of subjects was Serum zinc concentration 10019. 18–39 40–59 60 (years) (years) (years)p

Male Table 5 Multiple regression analysis (stepwise method) with serum zinc Morning fasting (mg/dl)a 86.8714.9b 81.3712.3b 76.079.8b concentration as the dependent variable Morning non-fasting (mg/dl)a 90.9714.6 78.878.9 68.1710.7 Afternoon (mg/dl)a 69.579.4 68.5710.5 62.5710.8 Variables b P-value

Female Time of blood sampling À0.405 o0.001 Morning fasting (mg/dl)a 83.4714.2b 82.878.8b 72.0711.0b Age À0.335 o0.001 Morning non-fasting (mg/dl) 76.1710.0 70.878.9 69.2711.7 Meal intake on the day À0.159 o0.001 Afternoon (mg/dl)a 68.2710.5 66.079.1 63.178.1 Total number of subjects was 1009. Total number of subjects was 1009 (392 male and 617 female). Table 2 R2 ¼ 0.360, Enter: Pp0.050, Remove: PX0.100 indicates the details of each number of subgroups. Stepwise regression analysis was applied to reveal the relationship between aSignificant difference in serum zinc among age groups, Po0.001 (ANOVA). serum zinc and confounding variables (time of blood sampling, meal intake on bSignificant difference in serum zinc among time/fasting status in each age the day, age and gender). Three variables were entered into the model and gender group (each of six groups), Po0.001 (ANOVA). excluding gender.

European Journal of Clinical Nutrition Ratio of low serum zinc levels in elderly Japanese M Kogirima et al 378 120

100

80 g /dl) µ 60

40 Serum zinc concentration ( 20 40 60 80 100 Age (y) Figure 2 Correlation between age and serum zinc concentration. There were 443 subjects. To exclude the confounding factor of the blood sampling time, we performed a simple correlation analysis only on the subjects with blood sampling in the morning regarding the relation between serum zinc level and age.

120 (mainly soy bean), other vegetables, dairy products and total n = 26 animal foods. Although elderly people have smaller zinc intakes from total animal foods compared to younger people, there was a comparatively large zinc intake from 80 plant foods, fish and shellfish.

Discussion 40 y = 3.210x + 46.274

Serum zinc (µg/dl) 2 R = 0.271 Through this study, we found that the prevalence of zinc P < 0.01 deficiency was high in residents in Nagano Prefecture, a 0 central part of Japan. By age adjustment (reference Japanese 0.0 5.0 10.0 15.0 20.0 model population structure in 1985), the incidence of low Zinc intake (mg) serum zinc was 21.1% in adults. It was especially high in elderly people (37.9% in those of 60 years old or older vs Figure 3 Relationship between zinc intake and serum zinc levels in subjects aged 60 years or older. 16.9% in those of 18–59 years old). In the elderly population, there was correlation between zinc intake and serum zinc levels. As there are few epidemiological studies on serum zinc levels in the Japanese (Ohguri et al., 1996; Ohguri et al., each age group, 18–39 years old; R ¼ 0.035, NS, 40–59 years 1997), in this study, we evaluated serum zinc levels using the old; R ¼À0.128, NS, X60 years old; R ¼À0.499, Po0.001). low cutoff value determined by Hotz et al. (2003). As a result, The subjects whose blood was collected in the morning subjects aged 60 years or older showed a high ratio of low were classified into two groups with low (o65 mg/dl) and serum zinc. On the other hand, the serum zinc levels in high (X90 mg/dl) serum zinc levels in 2003. Serum zinc levels subjects aged 59 years or younger were not particularly low in 90 of them were measured again after 1 year, in 2004. compared to those observed in adults (Garcı´a et al., 2000; Among them, blood was collected in the morning from 50 Hotz et al., 2003). subjects (26–94 years old). The results of the 50 subjects were One of the reasons that subjects aged 60 years or older similar to the results in 2003, being 64.2713.3 mg/dl in the showed a high ratio of low serum zinc is, as shown in low serum zinc group and 89.2711.0 mg/dl in the high group Figure 2, that serum zinc levels decrease with age after 60 (Po0.001). years old. The surveyed area had a larger aged population Figure 3 shows the relationship between zinc intake and than the average Japanese population (Table 1). The Japanese serum zinc level in the subjects aged 60 years or older among aging index in 2003 was 135.8 (Census by the Ministry of 50 subjects. A correlation was observed in the elderly Public Management, Home Affairs, Posts and Telecommuni- subjects. There was no significant correlation between zinc cations Statistics Bureau in Japan). The rapid expansion of intake and serum zinc level in younger subjects (R ¼ 0.117, the aged population is especially serious in local regions. P ¼ 0.585). On the other hand, the zinc intake of this result was Table 6 shows the correlation between zinc intake from similar to that of the Japanese zinc intake in those aged 60 each food group and serum zinc level. In the subjects of 60 years or older (average Japanese vs these subjects; male 9.1 vs years or older, positive correlations were observed between 9.4 mg/day, female 7.7 vs 7.5 mg/day) (National Nutrition serum zinc levels and the zinc intake from potatoes, beans Survey in Japan, 2003).

European Journal of Clinical Nutrition Ratio of low serum zinc levels in elderly Japanese M Kogirima et al 379 Table 6 Relationship between the zinc intakes from each food groups and serum zinc level

Variables Means of zinc intake (mg) R

Overall o60 years X60 years Overall o60 years X60 years n ¼ 50 n ¼ 24 n ¼ 26 n ¼ 50 n ¼ 24 n ¼ 26

Cereals 2.73 3.02 2.47 0.040 À0.163 0.148 Nuts and seeds 0.12 0.19 0.05a 0.118 À0.146 0.057 Potatoes 0.10 0.12 0.07 0.302* À0.101 0.561** Beans 0.53 0.57 0.48 0.460** 0.341 0.480* Dark green and yellow vegetables 0.30 0.25 0.35 0.097 À0.072 0.290 Other vegetables 0.44 0.41 0.48 0.230 0.224 0.391* Mushrooms 0.07 0.08 0.06 0.197 0.102 0.094 Sea plants 0.06 0.06 0.06 À0.165 0.044 À0.204 Fruits 0.10 0.09 0.11 0.040 À0.163 0.342 Fish and shellfish 0.70 0.65 0.75 0.179 À0.010 0.208 Meats 1.07 1.23 0.92 0.263 0.114 0.278 Eggs 0.39 0.36 0.41 À0.052 À0.301 0.235 Dairy products 0.70 1.02 0.41b 0.405** À0.159 0.553** Sweets and snack 0.12 0.20 0.04a 0.335* À0.024 0.347 Animal foods 2.86 3.27 2.48a 0.388** À0.020 0.491*

Mean values significantly different from that of under 60 years old; aPo0.05, bPo0.001. Significant correlation; *Po0.05, **Po0.01.

Of the subjects aged 18–59 years, 16.9% had low serum between serum zinc levels and zinc intake from potatoes, zinc. In this age population, there was no correlation beans, other vegetables, dairy products and total animal between zinc intake and serum zinc levels. The cause of foods. Animal foods such as meat are more popular sources low serum zinc remains to be studied in this population. of zinc than other plant foods in the West. However, for Hotz et al. (2003) determined the cutoff values after elderly Japanese people, various plant foods may be excluding subjects (11.9%) with high white blood cell important as sources of zinc, because meat intake is generally counts, current pregnancy or lactation and hormone use. low (Ministry of Health, Labor and Welfare, 2005). Because only serum zinc levels were measured in this study, A reduction in dietary intake may be another factor we might have overestimated the number of subjects with affecting low serum zinc levels in the elderly population low serum zinc levels. because energy intake (R ¼À0.433, Po0.01), protein intake The percentage of subjects with low serum zinc levels were (R ¼À0.363, Po0.05), fat intake (R ¼À0.536, Po0.001) and different between genders, as shown in Figure 1, which we meat intake (R ¼À0.303, Po0.05) negatively correlated with consider to be owing to aging. Serum zinc level decreases age in our study. A negative correlation was also observed with age (Figure 2) and Japanese life expectancy in males is between age and zinc intake in these subjects (R ¼À0.413 shorter than that in females (78.64 vs 85.59 years; an Po0.01). We believe that zinc intake decreases owing to the announcement of the Japanese Ministry of Health, Labour decline in total meal quantity, and that meat intake is and Welfare in 2006). Therefore, we surmise that the associated with age (National Nutrition Survey in Japan, decreasing of serum zinc level in males appears earlier than 2003). females. Regarding the effect of zinc intake on this There was a positive correlation between the zinc intake phenomenon, we need to conduct a large-scale dietary from beans and serum zinc levels. Beans generally contain a survey. large amount of phytic acid, which inhibits the absorption of Other studies have indicated the time of blood sampling, zinc. The bean dishes eaten by the subjects in this study were food intake (fasting or non-fasting), age and sex to be mainly tofu (soybean curd), natto (fermented soybeans) and confounding factors that affect serum zinc levels (Rea, 1989; miso (soybean paste), all of which are processed soybean Hotz et al., 2003). Because the time of blood sampling has a foods. These foods contain less phytic acid than raw soybean great impact on serum zinc levels (Table 5), dietary analysis does (Ohtsuki et al., 2001). Considering Japanese dietary was conducted only in the subjects whose blood samples habits, processed soybean foods are an essential source of were collected in the morning. zinc. The serum zinc levels in subjects aged 60 years or older The beneficial effects of zinc supplementation and zinc- positively correlated with the zinc intake from some food fortified diets have been reported in younger people groups. Therefore, by increasing such foods, elderly people (Roman˜a et al., 2005). Although there are few reports (Hyun may be able to recover their serum zinc levels. In people aged et al., 2004) on the effect of zinc intake in elderly people, our 60 years or older, positive correlations were observed results suggest that the supply of zinc through meals in

European Journal of Clinical Nutrition Ratio of low serum zinc levels in elderly Japanese M Kogirima et al 380 elderly people has a beneficial effect. In order to contribute Brown KH, Peerson J, Allen LH (1998). Effect of zinc supplementa- towards a healthy life, it is important to assay serum zinc tion on children’s growth: a meta-analysis of intervention trials. levels for early screening and to improve zinc nutritional Bibl Nutr Dieta 54, 76–83. Caulfield LE, Zavaleta N, Figueroa A (1999). Maternal zinc supple- status by diet or temporary supplementation in elderly mentation does not affect size at birth or pregnancy duration in people (Takeda et al., 2004; Andriollo-Sanchez et al., 2005; Peru. J Nutr 129, 1563–1568. Kurasawa and Kubori, 2006). Ervin RB, Stephenson JK (2002). Mineral intakes of elderly We also analyzed the relationship between serum zinc adult supplement and non-supplement users in the Third National Health and Nutrition Examination Survey. J Nutr 132, levels and the subjective symptoms associated with slight or 3422–3427. moderate zinc deficiency. Although only a few subjects were Garcı´a MJ, Alegrı´a A, Barber R, Farr R, Lagarda MJ (2000). Selenium, aware of symptoms associated with moderate zinc defi- copper, and zinc indices of nutritional status. Biol Trace Element Res ciency, our analysis showed that serum zinc levels related to 73, 77–83. Gibson RS, Yeudall F, Drost N (1998). Dietary interventions to the degree of appetite and tiredness. It has been reported that prevent zinc deficiency. Am J Clin Nutr 68, 484S–487S. zinc deficiency lowers gestation and affects the nervous Goldenberg RL, Tamura T, Neggers Y (1995). The effect of zinc system, thus reducing appetite (Levenson, 2003). Many supplementation on pregnancy outcome. JAMA 274, 463–468. important enzymes and cytokines require zinc to function Hambidge M, Cousins RJ, Costello RB (2000). Zinc and health: current status and future directions. JNutr130, 1341S–1520S. (Shankar and Prasad, 1998; Iskra and Majewski, 2000; McCall Hetland O, Brubakk E (1973). Diurnal variation in serum zinc et al., 2000; Klein et al., 2002), and it is considered that zinc concentration. Scand J Clin Lab Invest 32, 225–226. deficiency may adversely affect various vital functions and Hill T, Meunier N, Andriollo-Sanchez M, Ciarapica D, Hininger- reduce metabolism and daily activity. Favier I, Polito A et al. (2005). The relationship between the zinc nutritive status and biochemical markers of bone turnover in older In conclusion, when we observed our results with lower European adults: the ZENITH study. Eur J Clin Nutr 59 (Suppl 2), cutoffs (2.5th percentile) than the US, the zinc deficiency of S73–S78. the Japanese was high. The percentage of low serum zinc Hodkinson CF, Kelly M, Coudray C, Gilmore WS, Hannigan BM, levels rose with age, reaching 37.9% at 60 years old or older. O’Connor JM et al. (2005). Zinc status and age-related changes However, even in subjects aged 60 years old or older, the in peripheral blood leukocyte subpopulations in healthy men and women aged 55–70 y: the ZENITH study. Eur J Clin Nutr 59 serum zinc levels were high as the zinc intake was high. (Suppl 2), S63–S67. Hopley C, Salkeld G, Wang JJ, Mitchell P (2004). Cost utility of screening and treatment for early age related macular degenera- Acknowledgements tion with zinc and antioxidants. Br J Ophthalmol 88, 450–454. Hotz C, Peerson JM, Brown KH (2003). Suggested lower cutoffs of serum zinc concentrations for assessing zinc status: reanalysis of This study was conducted with the support of residents and the second National Health and Nutrition Examination Survey local government officials in Tomi City, Nagano Prefecture, data (1976–1980). Am J Clin Nutr 78, 756–764. and Mimaki Social Welfare Group. The dietary survey was Hyun TH, Barrett-Conner E, Milne DB (2004). Zinc intakes and plasma concentrations in men with osteoporosis: the Rancho supported the dietitians, Shigeko Takano and Naomi Tagawa. Bernardo Study. Am J Clin Nutr 80, 715–721. Fancl Corporation provided funding for this study. We wish Iskra M, Majewski W (2000). Copper and zinc concentrations and the to express our sincere appreciation to all those who assisted activities of ceruloplasmin and superoxide dismutase in athero- in this study. sclerosis obliterans. Biol Trace Element Res 73, 55–65. Klein C, Sunahara RK, Hudson TY, Heyduk T, Howlett AC (2002). Zinc inhibition of cAMP signalling. J Biol Chem 227, 11859–11865. Kurasawa R, Kubori S (2006). Zinc deficiency and its clinical features References in the cases found in Kitamimaki, a rural area in Japan. Biomed Res Trace Elements 17, 91 (abstract). Aggett PJ (1989). Severe zinc deficiency. In: Mills CF (ed). Zinc in Kurasawa R, Kubori S, Kamioka H, Okada S, Matsumura O (2005). human biology. New York: Springer-verlag, pp. 259–279. Zinc deficiency in aged residents and lowered serum-zinc- Andriollo-Sanchez M, Hininger-Favier I, Meunier N, Toti E, Zaccaria concentration as population level in the same rural region. Biomed M, Brandolini-Bunlon M et al. (2005). Zinc intake and status in Res Trace Elements 16, 60–64 (in Japanese). middle-aged and older European subjects: the ZENITH study. Eur J Lee DH, Anderson KE, Harnack LJ, Folsom AR, Jacobs Jr DR (2004). Clin Nutr 59 (Suppl 2), S37–S41. Heme iron, zinc, alcohol consumption, and colon cancer: Iowa Bates CJ, Evans PH, Dardenne M (1993). A trial of zinc-supplementa- Women’s Health Study. J Natl Cancer Inst 96, 403–407. tion in young rural Gambian children. Br J Nutr 69, 243–255. Levenson CW (2003). Zinc regulation of food intake: new insights on Bhutta ZA, Black RE, Brown KH (1999). Prevention of diarrhea and the role of neuropeptide Y. Nutr Rev 61, 247–258. pneumonia by zinc supplementation in children in developing Ma J, Betts NM (2000). Zinc and copper intakes and their major food countries: pooled analysis of randomized controlled trials. J Pediatr sources for older adults in the 1994–96 Continuing Survey of Food 135, 689–697. Intakes by Individuals (CSFII). J Nutr 130, 2838–2843. Black MM, Baqui AH, Zaman K, Persson LA, Arifeen SE, Le K et al. McCall KA, Huang C, Fierke CA (2000). Function and mechanism of (2004). Iron and zinc supplementation promote motor develop- zinc metalloenzymes. JNutr130, 1437S–1446S. ment and exploratory behavior among Bangladeshi infants. Am J McMillan EM, Rowe DJF (1982). Clinical significance of diurnal Clin Nutr 80, 903–910. variation in the estimation of plasma zinc. Clin Exp Dermatol 7, Briefel RR, Bialostosky K, Stephenson JK, McDowell MA, Ervin RB, 629–632. Wright JD (2000). Zinc intake of the U.S. population: Findings Ministry of Health, Labor and Welfare (2005). The National Nutrition from the Third National Health and Nutrition Examination Survey in Japan, 2003 (in Japanese). Dai-ichi Shuppan: Tokyo, Survey, 1988–1994. J Nutr 130, 1367S–1373S. pp. 162.

European Journal of Clinical Nutrition Ratio of low serum zinc levels in elderly Japanese M Kogirima et al 381 Ohguri M, Imaki M, Kawabata K, Irimajiri K, Tanada S (1997). Serum Peruvian children consuming wheat products fortified with zinc in relation to clinical parameters in human subjects. J Anal Bio iron only or iron and 1 of 2 amounts of zinc. Am J Clin Nutr 81, Sci 20, 235–241. 637–647. Ohguri M, Imaki M, Kawabata K, Yoshida Y, Nakamura T, Tanada S Shankar AH, Genton B, Baisor M (2000). The influence of zinc (1996). Difference of serum metals in relation to age and sex in supplementation on morbidity due to Plasmodium falciparum: human subjects. Jpn J Clin Ecol 5, 72–78. a randomized trial in pre-school children in Papua New Guinea. Ohtsuki K, Sato K, Nakamura Y (2001). Utilization of phytase in Am J Trop Med Hyg 62, 663–669. koji-natto: hydrolysis of phytate in soy foods and improvement Shankar AH, Prasad AS (1998). Zinc and immune function: the of the utilities of micro-amount minerals. Soy Protein Res, Japan 4, biological basis of altered resistance to infection. Am J Clin Nutr 68, 33–38 (in Japanese). 447S–463S. Osendarp SJM, van Raaij JMA, Darmstadt GL, Baqui AH, Hautvast Takeda N, Takaoka T, Ueda C, Toda N, Kalubi B, Yamamoto S (2004). JGAJ, Fuchs GJ (2001). Zinc supplementation during pregnancy Zinc deficiency in patients with idiopathic taste impairment with and effects on growth and morbidity in low birthweight infants: a regard to angiotensin converting enzyme activity. Auris Nasus randomised placebo controlled trial. Lancet 357, 1080–1085. Larynx 31, 425–428. Prasad AS (1991). Discovery of human zinc deficiency and studies in Yamamoto T (1996). Normal and disorder in taste development. an experimental human model. Am J Clin Nutr 53, 403–412. In brain and taste. (in Japanese). Kyouritsu Shuppan: Tokyo, Rea IM (1989). Sex and age changes in serum zinc levels. Nutr Res 9, pp. 195–217. 121–125. Yokoyama K, Araki S, Sato H, Aono H (2000). Circadian rhythms of Roman˜a DL, Salazar M, Hambidge KM, Penny ME, Peerson JM, Krebs seven heavy metals in plasma, erythrocytes and urine in men: NF et al. (2005). Longitudinal measurements of zinc absorption in observation in metal workers. Ind Health 38, 205–212.

European Journal of Clinical Nutrition