![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
3 Kenya Medical Research Institute, Centre for Public Health Research, Nairobi, 00202 Kenya; 4 Northern Ireland Centre for Food and Health, University of Ulster, Coleraine BT52 1SA, UK; and 5 HIV/AIDS Section, UNICEF NYHQ, New York, NY 10017
* To whom correspondence should be addressed. E-mail: di.thurnham{at}ulster.ac.uk.
 |
ABSTRACT |
|---|
 TOP ABSTRACT IntroductionMethodsResultsDiscussionLITERATURE CITED |
|---|
1-acid glycoprotein (AGP) to adjust hemoglobin and ferritin concentrations to clarify interpretation of iron status. Apparently healthy adults who tested positive twice for HIV but who had not reached stage IV or clinical AIDS were randomly allocated to receive a food supplement (n = 17 and 21) or the food plus a micronutrient capsule (MN; 10 men and 34 women, respectively) containing 30 mg iron/d. Hemoglobin, ferritin, CRP, and AGP concentrations were measured at baseline and 3 mo and subjects were divided into 4 groups (reference, no inflammation; incubating, raised CRP; early convalescence, raised AGP and CRP; and late convalescence, raised AGP). Correction factors (the ratios of the median for the reference group over each inflammatory group) improved the consistency of the ferritin but not the hemoglobin results. After correction, ferritin (but not hemoglobin) increased in both men (48 µg/L; P = 0.02) and women (12 µg/L; P = 0.04) who received MN but not in the food-only group. However, hemoglobin did improve in subjects who showed no inflammation both at baseline and mo 3 (P = 0.019), but ferritin did not increase in this group. In conclusion, ferritin concentrations were more closely linked to current inflammation than hemoglobin; hence, correction by inflammation biomarkers improved data consistency. However, low hemoglobin concentrations were the consequence of long-term chronic inflammation and improvements in response to MN supplements were only detected in subjects with no inflammation.
|
Introduction |
|---|
|
TOPABSTRACTIntroductionMethodsResultsDiscussionLITERATURE CITED |
|---|
1-acid glycoprotein (AGP), to account for inflammation (1). Blood ferritin concentration is a biomarker of iron storage and, currently, the WHO defines iron deficiency by ferritin concentrations <12 µg/L in children <5 y and <15 µg/L in those 
5 y. However, ferritin concentrations increase in the presence of an acute phase response and thus iron deficiency is potentially underestimated in developing countries where prevalence of inflammation is high (2,3). Excluding samples from an analysis where evidence of inflammation is present can potentially bias results and can also be impracticable where the prevalence of inflammation is extensive, as, for example, in a recent study in Gambia where more that two-thirds of infants at 12 mo had elevated CRP and AGP (4).
During acute inflammation, the normal control of iron metabolism is reorganized by the cytokines tumor necrosis factor , interleukin-6, and interleukin-1, the primary mediators of the acute phase response and in both animal (5) and human studies (6), plasma ferritin concentrations increase rapidly and in parallel with CRP (6). These data confirmed earlier reports that plasma ferritin concentrations paralleled those of CRP during tuberculosis, rheumatoid arthritis, neutropenic sepsis, etc., suggesting that ferritin responds as an early APP (7); however, the degree to which ferritin increased was influenced by the underlying iron status (7), whereas CRP was more related to the severity of disease (8). However, when patients recover from disease, there is a sharp decrease in CRP (9) concentrations that is not paralleled by a decrease in plasma ferritin (10), possibly because the half-life of plasma CRP is 5–7 h (9), whereas ferritin is closer to 30 h (11). Plasma AGP concentrations also remain elevated as clinical symptoms disappear (12,13); thus, during convalescence, plasma ferritin concentrations behave more like AGP.
Various methods have been used to interpret iron status in the presence of inflammation. Researchers have suggested using higher cut-offs for ferritin than the WHO criteria (14,15), multiple hematological criteria (16), ferritin:serum transferrin receptor ratios (17,18), and regression analysis against CRP, AGP, or erythrocyte sedimentation ratio (15,19). For a variety of reasons, none of these methods is entirely satisfactory. To some extent, the results from regression analysis can be predicted, because although ferritin remains elevated while both CRP and AGP are increased, the degree of elevation of ferritin will be different depending on which of the 2 APP is increased. High CRP and AGP concentrations represent different phases in the inflammatory response and will have a different relationship with ferritin. We have described elsewhere how the different phases of the inflammatory process in apparently healthy subjects can be defined to calculate specific correction factors (CF) for the different phases and we used these to correct plasma retinol concentrations (2). In this article, we apply the same principle to correct plasma ferritin and hemoglobin concentrations before and after iron intervention to determine whether removing the influence of inflammation assists in interpreting whether micronutrient supplementation improved iron status.
|
Methods |
|---|
|
TOPABSTRACTIntroductionMethodsResultsDiscussionLITERATURE CITED |
|---|
-tocopheryl acetate (6.71 mg RRR--tocopherol equivalents), vitamin C (70 mg), thiamin (1.4 mg), riboflavin (1.4 mg), niacin (18 mg), pyridoxine hydrochloride (1.9 mg), cyanocobalamin (2.6 mg), folic acid (400 µg), zinc (15 mg), copper (2 mg), selenium (65 mg), and iodine (150 µg). A blood sample was taken from persons who felt well enough. Persons who felt unwell were given an alternative date for blood sampling, but 6 subjects failed to return. A total of 174 persons provided a blood sample before any supplements were received. Insufficient blood or laboratory problems prevented some analyses and the results for 163 samples (56 men and 107 women) were the total number obtained at baseline. There were a large number of subjects who failed to complete the study and only 82 subjects provided another blood sample at 3 mo. Ethical approval for the study was obtained from the Scientific Steering Committee of the Kenya Medical Research Institute, the National Ethical Review Board, and the Contracts Review Committee of UNICEF East and Southern Africa Regional Office.
Biochemistry. Hemoglobin was determined by the cyanmethemoglobin method on 200 µL of freshly collected blood in local laboratories. Anemia was reported if hemoglobin was <110 g/L in men or <100 g/L in women (21). Plasma was prepared in the local laboratories and shipped frozen to Nairobi where it was stored at –20°C for up to 15 mo prior to analysis. Plasma ferritin concentrations were measured at the Biochemistry Department, Kenyatta National Hospital, Nairobi. Ferritin concentration was measured by an automated 1-step sandwich enzyme immunoassay with final fluorescence detection (Mini Vidas, BioMerieux). Kits provided manufacturers' standards and controls and a precision between 4 and 7% was obtained.
At the end of the study, samples were shipped in dry ice to Northern Ireland and stored at –70°C for up to an additional 2 mo before analysis. AGP and CRP were measured using DAKO reagents (Dako) using a Hitachi 912 Clinical Analyzer (Roche Diagnostics). Analysis required <100 µL and plasma was diluted automatically prior to analysis. Interassay precision for each of the APP was 
5%. Limits of detection for the APP were 
0.2 g/L for AGP and 0.1 mg/L for CRP. Inflammation was indicated when plasma concentration of CRP was >5 mg/L and AGP was >1.0 g/L (2).
Data handling.
The purpose of the analysis was to determine whether correcting plasma ferritin and hemoglobin concentrations for the influence of current inflammation would assist in detecting an improvement in iron status. The background to the method of correction is described elsewhere (2,22) (illustrative data in Table 1). CF are the ratios of the medians of the first, or reference, group to the respective inflammatory groups. Reference group subjects were characterized by having normal AGP (1.0 g/L) and normal CRP (5 mg/L) activity. The second group is described as being in the incubation phase or early stage of infection when CRP is rising but AGP is still within the normal range. The other 2 groups are early convalescence, when CRP and AGP are both elevated, and late convalescence, when only AGP is elevated. To correct hemoglobin and ferritin concentrations, CF were determined from the acute phase status both pre and post intervention and analytes were then corrected by multiplying by the appropriate CF. An abstract of the ferritin data was published previously (23).
|
|
Results |
|---|
|
TOPABSTRACTIntroductionMethodsResultsDiscussionLITERATURE CITED |
|---|
|
The relative similarity of the recruits in inflammatory status at the 2 time points was reflected also in the CRP and AGP concentrations (Table 3). Neither CRP nor AGP changed between baseline and 3 mo for either sex or treatment group. Furthermore, only 1 difference emerged in the comparisons between the subjects in the different treatment groups. Plasma AGP concentrations were higher in men who received the MN treatment than in those who received food at baseline (P = 0.046) and tended to be higher at 3 mo (P = 0.08). CRP concentrations of the men did not differ at baseline or 3 mo.
|
|
|
We also investigated whether the absence of any increase in hemoglobin in response to the MN supplement was due to the presence or absence of inflammation over the 3-mo intervention. The number of subjects in reference group 1 at both baseline and 3 mo did not differ (Table 2). However, in the reference group of the follow-up sample, 2 men and 5 women had previously been in the other inflammation groups and 1 hemoglobin result was lost, but the majority of subjects (altogether 80%) tended to remain in either the reference group (n = 31) or within the groups with inflammation (n = 36). The data were therefore reanalyzed according to whether subjects had always been in the reference group or were never in the reference group (Table 6). Both hemoglobin and ferritin concentrations differed between the sexes (repeated measures ANOVA; P < 0.001), but there was no interaction between sex and treatment group, so men and women were combined and the reference and combined inflammatory groups were analyzed separately.
|
There were no correlations between baseline or follow-up hemoglobin concentrations with CRP or AGP for subjects in either of the specific reference or inflammation subgroups in Table 6 (data not shown). However, hemoglobin concentrations of the reference subgroup were correlated with erythrocyte sedimentation rate (ESR) at both baseline (r –0.68; P < 0.001; n = 31) and follow-up (r –0.36; P = 0.05; n = 30). We previously showed that baseline ferritin concentrations for men and women combined (n = 163) were correlated with AGP (r = 0.595; P < 0.01), CRP (r = 0.303; P < 0.01), and ESR (r = 0.17; P < 0.05) (22); however, ferritin concentrations of subjects in the reference group and corrected plasma ferritin concentrations of all subjects combined did not correlate with AGP, CRP, or ESR (data not shown), suggesting that that correction removed the effects of inflammation from ferritin.
|
Discussion |
|---|
|
TOPABSTRACTIntroductionMethodsResultsDiscussionLITERATURE CITED |
|---|
The reasons for correcting the data were to remove the influence of inflammation from inflammation-sensitive biomarkers and avoid discarding data. Ferritin is particularly sensitive to the effects of inflammation as illustrated by the high median concentrations during both early and late convalescence (Tables 1 and 2). Ferritin concentrations rise rapidly with increased CRP (6) during the incubation phase, remain elevated while CRP is elevated (7) (i.e. sickness), and continue to remain high after the decline in CRP (10,24) (i.e. convalescence). The objective of correcting the ferritin data was to reduce all values in a particular inflammatory group by a factor that represented the difference between the medians of the reference and the particular acute phase group so that their new medians were the same as that of the reference group. We assumed that the ferritin concentrations in the reference group were representative of the healthy community. For men and women in the reference group in this study, median ferritin concentrations at baseline were 161 and 41 µg/L, respectively, and 145 and 53 µg/L at follow-up. In a separate study in the same area, ferritin concentrations for subjects without HIV were 84 µg/L in men (n = 164) and 30 µg/L in women (n = 968) (22,25), so correction may not have completely removed the influence of inflammation from plasma ferritin concentrations.
However, baseline ferritin or hemoglobin concentrations of subjects in the reference groups or corrected data for all subjects were not correlated with either AGP or CRP. In addition, although we previously showed that ESR was elevated in the reference group (22), neither uncorrected or corrected ferritin concentrations correlated with ESR at either baseline or follow-up, which indicated that plasma ferritin concentrations were not associated with that inflammatory component.
We calculated CF for the sexes separately, although the increase in ferritin concentration that accompanied inflammation was linked to both the inflammatory state (6) and the preinflammation ferritin concentration in the blood, i.e. iron status (7). If the change in concentration responded to both factors proportionately, then the CF should be the same for both men and women and similar CF were obtained in the specific inflammation groups. However, we used sex-specific CF and in addition we used the maximum amount of data available at the time of the examination. So baseline CF were calculated separately for 56 men and 107 women and the CF used for the follow-up data were obtained from 27 men and 55 women or 52 women in the case of hemoglobin. It could be argued that to calculate the baseline CF, we should have used the median values for the 27 men and 55 women, as done for the follow-up data. CF calculated from the baseline data in Table 2 are slightly different from those calculated from the complete baseline data, but the overall results were similar and we decided to use the larger data set to determine the CF. However, even when we used all the baseline ferritin data to calculate CF, there were only small numbers of subjects in the incubation and late convalescent groups, which meant that reliability of the CF for some inflammatory groups was poor. Nevertheless, the median ferritin concentrations obtained in the respective inflammatory groups conformed with expectations, i.e. highest concentrations were in the early convalescent group and lower concentrations were in the other groups. We therefore accepted the possible unreliability of some CF and used all 6 values to correct data irrespective of sample size, but an ongoing meta-analysis may simplify this procedure.
The WHO/CDC working group report also recommended hemoglobin should be measured in iron intervention studies (1). Anemia is an end-result of chronic inflammation (26), but the inhibition of erythropoiesis does not have immediate effects on blood hemoglobin concentration, as the life span of the RBC buffers against any effects of short infective episodes. It is prolonged and chronic inflammation that depresses hemoglobin; thus, correcting for APP concentrations in this study, which are indicators of current inflammation, did not affect hemoglobin concentrations (Tables 4 and 5) and there were no correlations between the APP and hemoglobin (data not shown). There were also no correlations between baseline hemoglobin and ferritin concentrations (not shown).
We therefore examined hemoglobin responses in reference group subjects separately from those who were not free of inflammation at both time points (Table 6) and found that subjects who were always in the reference group had increased hemoglobin concentration (P = 0.019) in response to the MN supplement, but ferritin concentration did not increase in this group (P = 0.116). In contrast, in those who exhibited inflammation at both time points, there was increased ferritin concentration (P = 0.021) but hemoglobin concentration did not change (P = 0.242). We do not know if these subjects were exposed to inflammation throughout the 3 mo, but we do know that hemoglobin concentrations remained low in this group, whereas both corrected and uncorrected ferritin concentrations increased in those receiving the MN supplement. These data suggest that the iron supplement was being used by all the subjects who received it and that, in the absence of inflammation, normal hematological processes channeled the iron mainly to the synthesis of hemoglobin, whereas in the presence of inflammation, erythropoiesis was inhibited and the iron was channeled to the liver, raising plasma ferritin concentrations.
Beard et al. (19) recently examined the relationship between ferritin, CRP, and AGP in an attempt to identify which inflammatory biomarker better identified persons with spurious elevations in iron status. They used data from 2 studies, namely African American infants and Guatemalan school-age children. Both log AGP and CRP values were significantly correlated with plasma ferritin concentrations (r = 0.70), but the unexplained variance was large (>50%) and the relationships were of poor predictive value (<72%).
The relatively poor predictive properties of both CRP and AGP alone for ferritin may be related to the fact that increases in all 3 compounds are only partially related to each other. Following an inflammatory response, ferritin concentrations rise rapidly (6) and remain elevated after the clinical phase subsides (7,10,24). Therefore, in a survey of apparently healthy subjects, ferritin will tend to be better correlated with CRP in the incubation and early convalescence groups, whereas ferritin will correlate better with AGP values in the later phase of convalescence. The proportion of subjects within those groups will also differ depending on the disease exposure within a community. To consider 2 extreme examples, in malaria-endemic Papua New Guinea, the proportions of preschoolers in the incubation, early, and late convalescent groups were 44:17:2% (27), whereas in Pakistani preschool children living in a nonmalarious area, they were 0.2:10:51% (2,28). In the former study, the onset of new infections was frequent, as indicated by the high proportion in the incubation phase, and CRP may predict ferritin better than in the second scenario. In the Pakistani study, where frequency of infection was probably less common than in Papua New Guinea and convalescence appeared to last longer, AGP will probably be the better predictor. Neither CRP nor AGP can explain all aspects of the ferritin response, because the latter encompasses a longer time interval than either of the individual APP. However, combining the acute phase responses of CRP and AGP to categorize subjects' phase of inflammation enables both the time and degree of ferritin response to be accommodated as described in this article.
In conclusion, this study showed that how subjects responded to an iron supplement depended very much on their inflammatory status. Both hemoglobin and ferritin biomarkers were influenced by inflammation, but only ferritin concentrations were directly related to current inflammation. Thus, the CF approach enabled plasma ferritin concentrations to be used to demonstrate benefits from the iron supplement. In contrast, hemoglobin concentrations were not closely related to current inflammatory status but were more strongly influenced by long-term or frequent bouts of inflammation that gradually depress hemoglobin by what is known as anemia of chronic inflammation. Therefore, to demonstrate an effect of iron supplementation on hemoglobin, it is necessary to identify those with the least risk of inflammation in the community such as those subjects in the reference groups at the 2 time points who had increased hemoglobin.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Author disclosures: A. S. W. Mburu, D. I. Thurnham, D. L. Mwaniki, E. M. Muniu, F. Alumasa, and A. de Wagt, no conflicts of interest.
6 Abbreviations used: APP, acute phase protein; AGP, -1-acid glycoprotein; CF, correction factor; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; MN, subjects received food supplement plus micronutrients.
Manuscript received 27 June 2007. Initial review completed 15 August 2007. Revision accepted 27 December 2007.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTIntroductionMethodsResultsDiscussionLITERATURE CITED |
|---|
1. WHO, CDC. Assessing the iron status of populations. Geneva: WHO Press; 2004
2. Thurnham DI, McCabe GP, Northrop-Clewes CA, Nestel P. Effect of subclinical infection on plasma retinol concentrations and assessment of prevalence of vitamin A deficiency: meta-analysis. Lancet. 2003;362:2052–8.[Medline]
3. Brown KH, Lanata CF, Yuen ML, Peerson JM, Butron B, Lonnerdal B. Potential magnitude of misclassification of a population's trace element status due to infection: example from a survey of young Peruvian children. Am J Clin Nutr. 1993;58:549–54.
4. Darboe MK, Thurnham DI, Morgan G, Adegbola RA, Secka O, Solon J, Jackson S, Northrop-Clewes CA, Fulford TJ, et al. Effectiveness of the new IVACG early high-dose vitamin A supplementation scheme compared to the standard WHO protocol: a randomised controlled trial in Gambian mothers and infants. Lancet. 2007;369:2088–96.[Medline]
5. Kobune M, Kohgo Y, Kato J, Miyazaki E, Niitsu Y. Interleukin-6 enhances hepatic transferrin uptake and ferritin expression in rats. Hepatology. 1994;19:1468–75.[Medline]
6. Feelders RA, Vreugdenhil G, Eggermont AMM, Kuiper-Kramer PA, van Eijk HG, Swaak AJG. Regulation of iron metabolism in the acute-phase response: interferon-
and tumor necrosis factor- induce hypoferraemia, ferritin production and a decrease in circulating transferrin receptors in cancer patients. Eur J Clin Invest. 1998;28:520–7.[Medline]
7. Baynes RD, Bezwoda W, Bothwell TH, Khan Q, Mansoor N. The non-immune inflammatory response: serial changes in plasma iron, iron-binding capacity, lactoferrin, ferritin and C-reactive protein. Scand J Clin Lab Invest. 1986;46:695–704.[Medline]
8. Thompson D, Milford-Ward A, Whicher JT. The value of acute phase protein measurements in clinical practice. Ann Clin Biochem. 1992;29:123–31.[Medline]
9. Young B, Gleeson M, Cripps AW. C-reactive protein: a critical review. Pathology. 1991;23:118–24.[Medline]
10. Northrop-Clewes CA. Interpreting indicators of iron status during an acute phase response: lessons from malaria and HIV. Ann Clin Biochem. 2008;45:18–32.
11. Worwood M, Cragg SJ, Williams AM, Wagstaff M, Jacobs A. The clearance of 131I-human plasma ferritin in man. Blood. 1982;60:827–33.
12. Fleck A, Myers MA. Diagnostic and prognostic significance of acute phase proteins. In: Gordon AH, Koj A, editors. The acute phase response to injury and infection. Amsterdam: Elsevier Scientific Publishers; 1985. p. 249–71.
13. Stuart J, Whicher JT. Tests for detecting and monitoring the acute phase response. Arch Dis Child. 1988;63:115–7.
14. UNICEF, UNU, WHO. Iron deficiency anaemia. Assessment, prevention and control. A guide for programme managers. IDA Consultation, Geneva 1993. WHO/NHD/01.3:1–114. Geneva: WHO; 2001.
15. Witte DL. Can ferritin be effectively interpreted in the presence of the acute-phase response? Clin Chem. 1991;37:484–5.
16. Looker AC, Gunter EW, Johnson CL. Methods to assess iron status in various NHANES surveys. Nutr Rev. 1995;53:246–54.[Medline]
17. Cook JD, Flowers CH, Skikne D. The quantitative assessment of body iron. Blood. 2003;101:3359–63.
18. Lee EJ, Oh EJ, Park YH, Lee HK, Kim BK. Soluble transferrin receptor (sTfR), ferritin, and sTfR/log ferritin index in anemic patients with nonhematologic malignancy and chronic inflammation. Clin Chem. 2002;48:1118–21.
19. Beard JL, Murray-Kolb LE, Rosales FJ, Solomons NW, Angelilli ML. Interpretation of serum ferritin concentrations as indicators of total-body iron stores in survey populations: the role of biomarkers for the acute phase response. Am J Clin Nutr. 2006;84:1498–505.
20. Mburu ASW, Mwaniki DL, Thurnham DI, Selenje L, de Wagt A, Muniu EM, Friis H, Krarup HB. Effects of multi-micronutrient supplements and food aid rations on the nutritional status and health of HIV+ adults (MINIFAR). Report to UNICEF July 2004, NairobiKenyaCPHR-KEMRI, pp 1–209.
21. DeMaeyer EM, Dallman PR, Gurney JM, Hallberg L, Sood SK, Srikantia SG. Preventing and controlling iron deficiency anaemia through primary health care: a guide for health administrators and programme managers. Geneva: WHO; 1989.
22. Thurnham DI, Mburu ASW, Mwaniki DL, Muniu EM, Alumasa F, de Wagt A. Using plasma acute phase protein concentrations to interpret nutritional biomarkers in apparently-healthy HIV-1 seropositive Kenyan adults. Br J Nutr. In press 2008.
23. Thurnham DI, Mburu ASW, Mwaniki DL, Alumasa F, Muniu EM. Using acute phase proteins to correct plasma ferritin concentrations enables ferritin to be used to show benefits of iron-supplementation in apparently-healthy, HIV1+ adults. Conference Proceeding. Washington: International Life Sciences Institution; In press 2008.
24. Wieringa FT, Dijkhuizen MA, West CE, Northrop-Clewes CA, Muhilal. Estimation of the effect of the acute phase response on indicators of micronutrient status in Indonesian infants. J Nutr. 2002;132:3061–6.
25. Mwaniki DL, Omwega AM, Minui EM, Mutunga JN, Akelola R, Shako BR, Gotink MH, Pertet AM. Anaemia and status of iron, vitamin A and zinc in Kenya. The 1999 National Survey Report. 2001:1–221. Nairobi, Ministry of Health.
26. Thurnham DI, Northrop-Clewes CA. Infection in the etiology of anemia. In: Kraemer K, Zimmermann MB, editors. Nutritional anemia. Basel: Sight & Life Press; 2007.
27. Shankar AH, Genton B, Semba RA, Baisor M, Paino J, Tamja S, Adiguma T, Wu L, Rare L, et al. Effect of vitamin A supplementation on morbidity due to Plasmodium falciparum in young children in Papua New Guinea: a randomised trial. Lancet. 1999;354:203–9.[Medline]
28. Paracha PI, Jamil A, Northrop-Clewes CA, Thurnham DI. Interpretation of vitamin A status in apparently-healthy Pakistani children using markers of sub-clinical infection. Am J Clin Nutr. 2000;72:1164–9.
This article has been cited by other articles:
|
|
 |
|
  T. T. Nga, P. Winichagoon, M. A. Dijkhuizen, N. C. Khan, E. Wasantwisut, H. Furr, and F. T. Wieringa Multi-Micronutrient-Fortified Biscuits Decreased Prevalence of Anemia and Improved Micronutrient Status and Effectiveness of Deworming in Rural Vietnamese School Children J. Nutr., May 1, 2009; 139(5): 1013 - 1021. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
