Inflammation and infections such as malaria affect micronutrient biomarker concentrations and hence estimates of nutritional status. It is unknown whether correction for C-reactive protein (CRP) and α1-acid glycoprotein (AGP) fully captures the modification in ferritin concentrations during a malaria infection, or whether environmental and sociodemographic factors modify this association. Cross-sectional data from eight surveys in children aged 6–59 months (Cameroon, Cote d’Ivoire, Kenya, Liberia, Malawi, Nigeria and Zambia; n 6653) from the Biomarkers Reflecting Inflammation and Nutritional Determinants of Anaemia (BRINDA) project were pooled. Ferritin was adjusted using the BRINDA adjustment method, with values < 12 μg/l indicating iron deficiency. The association between current or recent malaria infection, detected by microscopy or rapid test kit, and inflammation-adjusted ferritin was estimated using pooled multivariable linear regression. Age, sex, malaria endemicity profile (defined by the Plasmodium falciparum infection prevalence) and malaria diagnostic methods were examined as effect modifiers. Unweighted pooled malaria prevalence was 26·0 % (95 % CI 25·0, 27·1) and unweighted pooled iron deficiency was 41·9 % (95 % CI 40·7, 43·1). Current or recent malaria infection was associated with a 44 % (95 % CI 39·0, 52·0; P < 0·001) increase in inflammation-adjusted ferritin after adjusting for age and study identifier. In children, ferritin increased less with malaria infection as age and malaria endemicity increased. Adjustment for malaria increased the prevalence of iron deficiency, but the effect was small. Additional information would help elucidate the underlying mechanisms of the role of endemicity and age in the association between malaria and ferritin.

The combined sandwich-ELISA (s-ELISA; VitMin Lab, Germany) and the Quansys Q-Plex™ Human Micronutrient Array (7-Plex) are multiplex serum assays that are used to assess population micronutrient status in low-income countries. We aimed to compare the agreement of five analytes, α-1-acid glycoprotein (AGP), C-reactive protein (CRP), ferritin, retinol-binding protein 4 (RBP4) and soluble transferrin receptor (sTfR) as measured by the 7-Plex and the s-ELISA. Serum samples were collected between March 2016 and December 2017. Pregnant women (n 249) were recruited at primary healthcare clinics in Johannesburg, and serum samples were collected between March 2016 and December 2017. Agreement between continuous measurements was assessed by Bland–Altman plots and concordance measures. Agreement in classifications of deficiency or inflammation was assessed by Cohen’s kappa. Strong correlations (r > 0·80) were observed between the 7-Plex and s-ELISA for CRP and ferritin. Except for CRP, the 7-Plex assay gave consistently higher measurements than the s-ELISA. With the exception of CRP (Lin’s ρ = 0·92), there was poor agreement between the two assays, with Lin’s ρ < 0·90. Discrepancies of test results difference between methods increased as the serum concentrations rose. Cohen’s kappa for all the five analytes was < 0·81 and ranged from slight agreement (vitamin A deficiency) to substantial (inflammation and Fe deficiency) agreement. The 7-Plex 1.0 is a research and or surveillance tool with potential for use in low-resource laboratories but cannot be used interchangeably with the s-ELISA. Further optimising and validation is required to establish its interchangeability with other validated methods.

We studied ethnic differences in terms of iron status during pregnancy between Dutch women and other ethnicities and explore to what extent these differences can be explained by environmental factors. This cross-sectional population-based study (2002–2006) was embedded in the Generation R study and included a total of 4737 pregnant women from seven ethnic groups (Dutch, Turkish, Moroccan, Cape Verdean, Surinamese-Hindustani, Surinamese-Creole and Antillean). Ethnicity was defined according to the Dutch classification of ethnic background. Ferritin, iron and transferrin were measured in early pregnancy. The overall prevalence of iron deficiency was 7 %, ranging from 4 % in both Dutch and Surinamese-Creoles, to 18 % in Turkish, Moroccan and Surinamese-Hindustani women. Iron overload was most prevalent in Surinamese-Creole (11 %) and Dutch (9 %) women. Socioeconomic factors accounted for 5–36 % of the differences. Income was the strongest socioeconomic factor in the Cape Verdean and Surinamese-Hindustani groups and parity for the Turkish and Moroccan groups. Lifestyle determinants accounted for 8–14 % of the differences. In all groups, the strongest lifestyle factor was folic acid use, being associated with higher iron status. In conclusion, in our population, both iron deficiency and iron overload were common in early pregnancy. Our data suggest that ethnic differences in terms of socioeconomic and lifestyle factors only partly drive the large ethnic differences in iron status. Our data support the development of more specific prevention programmes based on further exploration of socioeconomic inequities, modifiable risk and genetic factors in specific ethnic subgroups, as well as the need for individual screening of iron status before supplementation.

Cryo-electron microscopy, widely used for high-resolution protein structure determination, does not require staining. Yet negative staining with heavy metal salts such as uranyl acetate has been in persistent demand since the 1950s due to its image contrasting capabilities at room temperature with a common electron microscope. However, uranium compounds are nuclear fuel materials and are tightly controlled worldwide. Acetates of each lanthanoid series elements except promethium are prepared at the same concentration (2%(w/v)) and used as a model on horse spleen ferritin for electron microscopic analysis to systematically evaluate their effectiveness as electron staining reagents for the protein. Analysis shows that the triacetates of samarium and europium, followed by gadolinium and erbium, and then lanthanum and neodymium could function as electron staining reagents. Thulium-triacetate precipitates thin plate-like crystals and may be used for selecting better imaging fields. Of the 14 lanthanoid-triacetates examined, about half are viable alternatives to uranyl acetate as an electron staining reagent for ferritin, and there appears an optimal range in ionic sizes for promising lanthanoids. This is the first systematic investigation of lanthanoid transition heavy metal triacetates from the viewpoint of lanthanoid contraction, using density distribution histograms of electron micrographs as an indicator for comparison with uranyl acetate.

Inflammation and infections such as malaria affect estimates of micronutrient status. Medline, Embase, Web of Science, Scopus and the Cochrane library were searched to identify studies reporting mean concentrations of ferritin, hepcidin, retinol or retinol binding protein in individuals with asymptomatic or clinical malaria and healthy controls. Study quality was assessed using the US National Institute of Health tool. Random effects meta-analyses were used to generate summary mean differences. In total, forty-four studies were included. Mean ferritin concentrations were elevated by: 28·2 µg/l (95 % CI 15·6, 40·9) in children with asymptomatic malaria; 28·5 µg/l (95 % CI 8·1, 48·8) in adults with asymptomatic malaria; and 366 µg/l (95 % CI 162, 570) in children with clinical malaria compared with individuals without malaria infection. Mean hepcidin concentrations were elevated by 1·52 nmol/l (95 % CI 0·92, 2·11) in children with asymptomatic malaria. Mean retinol concentrations were reduced by: 0·11 µmol/l (95 % CI −0·22, −0·01) in children with asymptomatic malaria; 0·43 µmol/l (95 % CI −0·71, −0·16) in children with clinical malaria and 0·73 µmol/l (95 % CI −1·11, −0·36) in adults with clinical malaria. Most of these results were stable in sensitivity analyses. In children with clinical malaria and pregnant women, difference in ferritin concentrations were greater in areas with higher transmission intensity. We conclude that biomarkers of iron and vitamin A status should be statistically adjusted for malaria and the severity of infection. Several studies analysing asymptomatic infections reported elevated ferritin concentrations without noticeable elevation of inflammation markers, indicating a need to adjust for malaria status in addition to inflammation adjustments.

Iron deficiency (ID) in early life is associated with morbidities. Most fetal iron required for infant growth is acquired in the third trimester from maternal iron store. However, how prenatal iron level affects ferritin level in early infancy remains controversial. This study aimed to examine the associations between maternal ferritin levels and cord blood serum ferritin (CBSF) and to compare the ferritin levels between different feeding practices in early infancy. Healthy Chinese mothers with uncomplicated pregnancy and their infants were followed up at 3 months post-delivery for questionnaire completion and infant blood collection. Infants who were predominantly breastfed and those who were predominantly formula fed were included in this analysis. Serum ferritin levels were measured in maternal blood samples collected upon delivery, cord blood and infant blood samples at 3 months of age. Ninety-seven mother–baby dyads were included. Maternal ID is common (56 %) while the CBSF levels were significantly higher than maternal ferritin levels. Only three infants (3 %) had ID at 3 months of age. There were no significant correlations between maternal ferritin levels with CBSF (r 0·168, P = 0·108) nor with infant ferritin levels at 3 months of age (r 0·023, P = 0·828). Infant ferritin levels at 3 months were significantly and independently associated with CBSF (P = 0·007) and birth weight (P < 0·001) after adjusting for maternal age, parity, maternal education, infant sex and feeding practice. In conclusion, maternal ID was common upon delivery. However, maternal ferritin levels were not significantly associated with CBSF concentrations nor infant ferritin concentrations at 3 months of age.

Adequate iron supply in pregnancy is important for both the woman and the fetus, but iron status is often assessed late in first trimester, if assessed at all. Therefore, identification of factors associated with iron status is important to target vulnerable groups with increased risk of deficiency. Our objectives were to (1) describe iron status in mid-pregnancy and (2) identify sociodemographic and lifestyle predictors of pregnancy iron status. This cross-sectional study uses data from The Norwegian Mother, Father and Child Cohort Study (collected 2002–2008) and The Medical Birth Registry of Norway. Iron status was measured as non-fasting plasma ferritin (P-Fe) and transferrin in gestational week (GW) 18 (n 2990), and by lowest reported Hb in GW 0–30 (n 39 322). We explored predictors of iron status with elastic net, linear and log-binomial regression models. Median P-Fe was 33 μg/l, and 14 % had depleted iron stores (P-Fe <15 μg/l). P-Fe below 30 μg/l was associated with reduced Hb. We identified eleven predictors, with interpregnancy interval (IPI) and parity among the most important. Depleted iron stores was more common among women with IPI < 6 months (56 %) and 6–11 months (33 %) than among those with IPI 24–59 months (19 %) and among nulliparous women (5 %). Positively associated factors with iron status included hormonal contraceptives, age, BMI, smoking, meat consumption and multi-supplement use. Our results highlight the importance of ferritin measurements in women of childbearing age, especially among women not using hormonal contraceptives and women with previous and recent childbirths.

The present study aimed to (a) establish the frequency of consumption of red meat and eggs; (b) determine serum ferritin levels (μg/l); and (c) establish the relationship between serum ferritin and the consumption of red meat and eggs. In Colombia during 2014–2018, an analytical study was conducted in 13 243 Colombian children between the ages of 5 and 17 years, based on cross-sectional data compiled by ENSIN-2015 (Encuesta Nacional de la Situación Nutricional en Colombia-2015) on serum ferritin levels and dietary consumption based on a questionnaire of the frequency of consumption. Using simple and multiple linear regression, with the serum ferritin level as the dependent variable and the frequency of consumption as the main explanatory variable, the crude and adjusted partial regression coefficients (β) between serum ferritin levels and consumption were calculated. The frequency of habitual consumption of red meat was 0⋅49 (95 % CI 0⋅47, 0⋅51) times/d. The frequency of habitual egg consumption was 0⋅76 (95 % CI 0⋅74, 0⋅78) times per d. The mean serum ferritin level in men was 41⋅9 (95 % CI 40⋅6, 43⋅1) μg/l and in women, 35⋅7 (95 % CI 34⋅3, 37⋅7) μg/l (P < 0⋅0001). The adjusted β between the consumption of red meat and eggs and serum ferritin levels were β = 3⋅0 (95 % CI 1⋅2, 4⋅7) and β = 2⋅5 (95 % CI 1⋅0, 3⋅9) for red meat and eggs, respectively. In conclusion, red meat and eggs are determinants of serum ferritin levels in Colombia and, therefore, could be considered public policy options to reduce anaemia and Fe deficiency.

ADHD is one of the most common neurobehavioral disorders among children and adolescents. In this prospective study, we aimed to measure circulating zinc and ferritin levels in children with ADHD, pick up the deficient ones to give zinc and iron supplements then compare before and after treatment according to their Conner’s scores and Wecsler IQ test. Current study included fifty children diagnosed as having ADHD by DSMV criteria, their zinc and ferritin levels were measured by Colorimetric method and enzyme-linked immunosorbent assay (ELISA) respectively. They were divided into: group I (zinc only deficient),group II (zinc and ferritin deficient),group III (non-deficient), cases with mineral deficiency received zinc (55 mg/day) and/or iron (6 mg/kg/day) for 6 months then reassessed by parent Conner’s rating scale. In group 1, there was no significant difference between the Wecsler verbal and non-verbal IQ scores and oppositional and cognitive problems in Conner’s scores before and after zinc supplements, although there was significant improvement in attention, hyperactivity, emotional liability and impulsivity. In group II, there was significant improvement in verbal and total IQ but not in performance IQ, also there was significant improvement in hyperactivity, emotional liability and impulsivity with no significant difference in oppositional, cognitive problems and inattention before and after zinc/ iron supplements. In Conclusion, Zinc supplements in adjuvant to the main treatment significantly improved symptoms of ADHD children. However, a combined zinc and iron supplements was superior to zinc alone in alleviating ADHD symptoms as well as IQ improvement.

Associations between ferritin and insulin sensitivity have been described in recent studies. The possible association showed conflicting results by sex and menopausal status. We aimed to investigate the cross-sectional association of ferritin levels with insulin resistance and β-cell function. A total of 2518 participants (1033 men, 235 pre-menopausal women and 1250 post-menopausal women) were enrolled from the Changfeng Study. A standard interview was conducted, as well as anthropometric measurements and laboratory analyses, for each participant. The serum ferritin level was measured using electrochemiluminescence immunoassay. Insulin resistance and β-cell function indices were derived from a homeostasis model assessment. The results showed that the serum ferritin levels were 250·4 (sd 165·2), 94·6 (sd 82·0) and 179·8 (sd 126·6) ng/ml in the men, pre-menopausal and post-menopausal women, respectively. In fully adjusted models (adjusting for age, current smoking, BMI, waist:hip ratio, systolic blood pressure, diastolic blood pressure, TAG, HDL-cholesterol, LDL-cholesterol, log urine albumin:creatinine ratio, leucocytes, alanine aminotransferase, aspartate aminotransferase and γ-glutamyl transpeptidase), serum ferritin concentrations are significantly associated with insulin resistance in men and post-menopausal females, and the null association was observed in pre-menopausal females. Interestingly, an increased β-cell function associated with higher ferritin was observed in post-menopausal participants, but not in male participants. In conclusion, these results suggested that elevated serum ferritin levels were associated with surrogate measures of insulin resistance among the middle-aged and elderly male and post-menopausal women, but not in pre-menopausal women.

Childhood intake of animal foods is associated with age at first menstrual period (menarche). It is unknown whether the micronutrients present in these foods could explain this association. Our objective was to investigate the associations of micronutrient status biomarkers in middle childhood with age at menarche. We quantified circulating Hb, ferritin, mean corpuscular volume, Zn, vitamin B12, erythrocyte folate and retinol in 1464 pre-menarcheal girls aged 5–12 years in Bogotá, Colombia, and followed them for a median 5·7 years for the occurrence and date of menarche. We estimated median age at menarche and hazard ratios (HR) with 95 % CI by levels of each biomarker with use of Kaplan–Meier survival probabilities and Cox regression, respectively. Median age at menarche was 12·4 years. Middle childhood Hb was inversely related to age at menarche whereas plasma ferritin was positively associated with this outcome in a linear manner. HR of menarche for every 1 sd of Hb (11 g/l) and ferritin (23·2 µg/l) were 1·11 (95 % CI 1·04, 1·18; P=0·001) and 0·94 (95 % CI 0·88, 0·99; P=0·02), respectively, after adjustment for baseline age, C-reactive protein concentration, maternal age at menarche and parity and socioeconomic status. The association with ferritin was stronger in girls aged 9–10 years at baseline. Additional adjustment for baseline height- and BMI-for-age did not change the results. We conclude that higher Fe status in middle childhood is related to later age at menarche whereas Hb concentrations are inversely associated with age at onset of menses.

Previous studies have supported the theory that there is a positive association between ferritin and carotid atherosclerosis in Western people. Diet plays an important role in determining serum ferritin concentration. Asian dietary patterns are different from Western dietary patterns, implying that there may be a difference in the association of ferritin with carotid atherosclerosis between Asian and Western people. However, few studies focus on the association between ferritin and carotid atherosclerosis among Asians. The aim of this study was to investigate how serum ferritin levels are associated with carotid atherosclerosis in an Asian adult population. A cross-sectional assessment was performed in 8302 adults in Tianjin, China. Carotid intima-media thickness (IMT) and plaques were assessed using ultrasonography, and serum ferritin was measured using the protein chip-chemiluminescence method. Multiple logistic regression analysis was used to examine the association between quartiles of serum ferritin concentration and carotid atherosclerosis. In the present study, the overall prevalence of IMT and carotid plaques in participants is 29·2 and 22·7 %, respectively. In women, after adjustments for potentially confounding factors, the OR of IMT and carotid plaques by increasing serum ferritin quartiles were 1·00, 1·39 (95 % CI 0·98–1·99), 1·39 (95 % CI 0·99–1·97), 1·81 (95 % CI 1·30–2·55) (Pfor trend<0·001) and 1·00, 1·24 (95 % CI 0·89–1·73), 1·18 (95 % CI 0·85–1·65), 1·59 (95 % CI 1·15–2·20) (Pfor trend<0·01), respectively. However, no association was found between serum ferritin and carotid atherosclerosis in men. The study demonstrated that increased serum ferritin levels are independently associated with IMT and carotid plaques in Asian women but not in Asian men.

Current evidence of the relationship between diets and Fe status is mostly derived from studies in developed countries with Western diets, which may not be translatable to Chinese with a predominantly plant-based diet. We extracted data that were nationally sampled from the 2009 wave of China Health and Nutrition Survey; dietary information was collected using 24-h recalls combined with a food inventory for 3 consecutive days. Blood samples were collected to quantify Fe status, and log-ferritin, transferrin receptor and Hb were used as Fe status indicators. In total, 2905 (1360 males and 1545 females) adults aged 18–50 years were included for multiple linear regression and stratified analyses. The rates of Fe deficiency and Fe-deficiency anaemia were 1·6 and 0·7 % for males and 28·4 and 10·7 % for females, respectively. As red meat and haem Fe consumption differed about fifteen to twenty times throughout the five groups, divided by quintiles of animal protein intake per 4·2 MJ/d, only Fe status as indicated by log-ferritin (P=0·019) and transferrin receptor (P=0·024) concentrations in males was shown to be higher as intakes of animal foods increased. Log-ferritin was positively associated with intakes of red meat (B=0·3 %, P=0·01) and haem Fe (B=12·3 %, P=0·010) in males and with intake of non-haem Fe in females (B=2·2 %, P=0·024). We conclude that diet has a very limited association with Fe status in Chinese adults consuming a traditional Chinese diet, and a predominantly plant-based diet may not be necessarily responsible for poor Fe status.

The metabolic syndrome (MetS) is known to be associated with elevated serum ferritin levels. The possible association with other Fe markers has been less well studied. We aimed to investigate the cross-sectional association of soluble transferrin receptor (sTfR) and ferritin levels with the MetS components, insulin resistance and glycosylated Hb (HbA1C). The sample consisted of 725 adults, aged 19–93 years (284 men, 151 premenopausal and 290 postmenopausal women), from the Croatian island of Vis. Serum sTfR and ferritin levels were measured by immunoturbidimetry and electrochemiluminescence assays, respectively. The MetS was defined using modified international consensus criteria. Logistic and linear regression analyses were conducted to investigate the associations adjusting for age, fibrinogen, smoking status, alcohol consumption and BMI. Prevalence of the MetS was 48·7 %. Standardised values of ferritin were positively associated with all of the MetS components (except high blood pressure and waist circumference) in men (P<0·05). Ferritin was significantly associated with the MetS in men (adjusted OR 1·78 (95 % CI 1·31, 2·42)) and postmenopausal women (1·71 (95 % CI 1·12, 2·62)). Interestingly, sTfR was independently and positively associated with homoeostatic model assessment for insulin resistance in men (adjusted β=0·44 (95 % CI 0·14, 0·75), P=0·004) and postmenopausal women (adjusted β coefficient=0·34 (95 % CI 0·05, 0·63), P=0·020). However, there was no significant relationship between serum sTfR levels and the MetS or its components. Neither ferritin nor sTfR was significantly associated with HbA1C (P>0·05). sTfR levels could be spuriously elevated in subjects with insulin resistance and without association with the MetS or its components. We conclude that different markers of Fe metabolism are not consistently associated with cardiometabolic risk.

While prenatal Fe supplementation prevents maternal Fe deficiency and anaemia, it is uncertain whether it improves infant health outcomes, at least when taken by Fe-replete women. Inflammation as well as physiological changes complicates the assessment of Fe status during pregnancy. In the present study, we measured the concentrations of serum ferritin and soluble transferrin receptors (sTfR), Hb and the acute-phase proteins C-reactive protein (CRP) and α1-antichymotrypsin (ACT) in a cross-sectional study among 738 pregnant women attending antenatal care in Guinea-Bissau, West Africa. Multiple linear regression analysis was used to identify the predictors of Fe status markers. The mean gestational age was 23 (sd 7) weeks. Serum ferritin values were lower with progressing gestation, from 27 % lower during weeks 16–20 of gestation up to 59 % lower after 29 weeks of gestation compared with early pregnancy. Using cut-off values for Fe deficiency as established in non-pregnant individuals, 52 % of the women had sTfR levels >2·3 mg/l, while only 25 % had serum ferritin levels < 12 μg/l. Serum ferritin but also sTfR levels were elevated at increased serum CRP and ACT levels. The proportion of sTfR >2·3 mg/l decreased to 47 % after adjustment for elevated serum CRP and ACT levels. On the contrary, the proportion of serum ferritin < 12 μg/l increased to 33 % after adjustment for ACT and CRP. The high proportion of elevated serum sTfR calls for pregnancy-specific cut-offs since increased erythropoiesis is expected in response to increased plasma volume of pregnancy. The present study further underlines the need to adjust for inflammation when serum sTfR and serum ferritin are used to assess Fe status in pregnancy.

Postmenopausal women are at increased risk of CVD: the increased serum ferritin level may be involved in the pathogenesis. The aim of the present study is to investigate the relationship of ferritin and carotid atherosclerosis in postmenopausal women. A total of 1178 postmenopausal women (mean age, 60·8 years) were enrolled from the Changfeng Study. A standard interview, anthropometric measurements and laboratory analyses were performed for each participant. Bilateral CIMT (carotid intima–media thickness) were measured using ultrasonography, and the presence of carotid plaques was assessed. Serum ferritin was measured using electrochemiluminescence immunoassay. The results showed that serum ferritin was 181·9 (sd 65·8) ng/ml in the postmenopausal women. Multivariate, linear, stepwise regression analysis demonstrated that age (standardised β = 0·233, P< 0·001), alanine transaminase (standardised β = 0·194, P< 0·001), log homeostasis model assessment index for insulin resistance (standardised β = 0·181, P< 0·001), TAG (standardised β = 0·083, P= 0·003), Hb (standardised β = 0·080, P= 0·004) and PPG (2-h glucose levels following a 75-g oral glucose challenge) (standardised β = 0·079, P= 0·004) were independently associated with serum ferritin. Compared with the ferritin level of subjects in the first quartile, that in the fourth quartile had greater CIMT, and higher prevalence of carotid plaque. After adjusting for conventional CVD risk factors, Hb, leucocytes, log urine albumin:creatinine ratio and liver function, the ferritin level of postmenopausal women in the fourth quartile had a 1·587-fold increased risk of carotid plaques relative to those in the lowest quartile. In conclusion, these results suggest that serum ferritin is independently and positively associated with carotid atherosclerosis in postmenopausal women and that ferritin may be implicated in atherosclerosis.

Anaemia, redistribution of Fe, malnutrition and heightened systemic inflammation during HIV infection confer an increased risk of morbidity and mortality in HIV patients. We analysed information on Fe status and inflammation from a randomised, double blind, controlled phase-III clinical trial in Lusaka, Zambia and Mwanza, Tanzania. Malnourished patients (n 1815) were recruited at referral to antiretroviral therapy (ART) into a two-stage nutritional rehabilitation programme, randomised to receive a lipid-based nutrient supplement with or without added micronutrients. Fe was included in the intervention arm during the second stage, given from 2 to 6 weeks post-ART. Hb, serum C-reactive protein (CRP), serum ferritin and soluble transferrin receptor (sTfR) were measured at recruitment and 6 weeks post-ART. Multivariable linear regression models were used to assess the impact of the intervention, and the effect of reducing inflammation from recruitment to week 6 on Hb and Fe status. There was no effect of the intervention on Hb, serum ferritin, sTfR or serum CRP. A one-log decrease of serum CRP from recruitment to week 6 was associated with a 1·81 g/l increase in Hb (95 % CI 0·85, 2·76; P< 0·001), and a 0·11 log decrease in serum ferritin (95 % CI − 0·22, 0·03; P= 0·012) from recruitment to week 6. There was no association between the change in serum CRP and the change in sTfR over the same time period (P= 0·78). In malnourished, HIV-infected adults receiving dietary Fe, a reduction in inflammation in the early ART treatment period appears to be a precondition for recovery from anaemia.