Iron deficiency anaemia is of major concern in low-income settings, especially for women of childbearing age. Oral iron substitution efficacy is limited by poor compliance and iron depletion severity. We aimed to assess the efficacy and safety of intravenous ferric carboxymaltose versus oral iron substitution following childbirth in women with iron deficiency anaemia in Tanzania.
Between Oct 8, 2015, and March 14, 2017, 533 individuals were screened and 230 were enrolled and randomly assigned to a study group (114 to intravenous iron, 116 to oral iron). At 6 weeks, 94 (82%) participants in the intravenous iron group and 92 (79%) in the oral iron group were assessed for the primary outcome. 75 (80%) participants in the intravenous iron group and 47 (51%) in the oral iron group had normalised haemoglobin (odds ratio 4·65, 95% CI 2·33–9·27). There were two mild to moderate infusion-related adverse events; and five serious adverse events (three in the intravenous iron group, two in the oral iron group), unrelated to the study medication.
Intravenous iron substitution with ferric carboxymaltose was safe and yielded a better haemoglobin response than oral iron. To our knowledge, this is the first study to provide evidence of the benefits and safety of intravenous iron substitution in a low-income setting.
Vifor Pharma, R Geigy-Stiftung, Freiwillige Akademische Gesellschaft, and Swiss Tropical and Public Health Institute.
In Africa, anaemia is a major public health concern: approximately 60% of the population have anaemia
due to malnutrition, pregnancies, haemoglobinopathies, iron deficiency, or functional iron deficiency in chronic infectious diseases, negatively affecting quality of life and national socioeconomic status.
It is generally assumed that 50% of the cases of anaemia are due to iron deficiency.
Menstruating individuals are at high risk of iron deficiency anaemia, and this risk is further increased during pregnancy and unsafe delivery.
Post-partum anaemia adversely affects maternal recovery, cognition, and maternal–infant interactions.
Evidence before this study
We searched PubMed in 2012, and in September, 2019, for studies comparing the use of intravenous ferric carboxymaltose versus oral iron in the treatment of iron deficiency anaemia in Africa or low-income countries using the terms “anaemia”, “iron deficiency”, “treatment”, “low-income”, AND “Africa”. We did not find any studies reporting the use of intravenous ferric carboxymaltose in resource-limited settings. Iron deficiency prevention and treatment in resource-limited settings is done by oral nutritional iron fortification or iron tablets. A meta-analysis from 2017 included data from 14 studies in high-income settings and found a benefit of ferric carboxymaltose over oral iron in correcting anaemia and iron deficiency. Ferric carboxymaltose is known to be safer than the older intravenous iron preparations such as iron dextran or iron sucrose.
Added value of this study
To our knowledge, this is the first randomised trial to assess the efficacy and safety of intravenous ferric carboxymaltose compared with oral iron in the treatment of iron deficiency anaemia following childbirth in a resource-limited setting. We found that intravenous ferric carboxymaltose was more effective than oral iron therapy in correcting anaemia and iron deficiency, in line with results from high-income settings. We have shown that ferric carboxymaltose can be safely infused in district and local hospitals in a resource-limited setting.
Implications of all the available evidence
Use of intravenous iron in the peripartum period is safe, feasible, and provides better correction of iron deficiency and anaemia than oral iron in resource-limited settings where incidence is particularly high.
Oral iron substitution is cheap, but requires fasting intake for maximal effect and good compliance over a long period, which might be challenging given the common side effects of obstipation, diarrhoea, and abdominal pain. Compliance in resource-limited settings is dependent not only on tolerance to the medication and its clinical effectiveness, but also on socioeconomic and educational factors.
Compliance with oral iron substitution might therefore be worse in resource-limited than in high-income settings, thus jeopardising its effectiveness.
where this treatment is increasingly replacing oral iron substitution and intravenous iron substitution with iron dextran (which carries a risk of anaphylactic reactions) or iron sucrose (which has a different safety profile).
However, the different socioeconomic, cultural, and medical conditions in low-resource settings—including medication access, perception of medication needs, compliance, and the burden of concomitant diseases—might influence the effectiveness and safety of iron substitution modality compared with high-income countries. Therefore, the most effective treatment approaches for iron deficiency anaemia in different low-resource settings are not known. Our aim was to compare the safety and efficacy of intravenous iron substitution by ferric carboxymaltose versus oral iron substitution in post-partum women in a rural setting in Tanzania.
Women presenting to the hospital antenatal care services were screened by nurses of the ward under supervision of physicians or clinical officers of the Ifakara Health Institute. Eligible participants were close to delivery; had iron deficiency anaemia, defined as a haemoglobin concentration of less than 110 g/L and a ferritin concentration of less than 50 μg/L measured within 14 days before childbirth; lived close to the hospital; and agreed to attend scheduled follow-up visits. Participants were excluded if they were HIV positive; had a known haemoglobinopathy; had a C-reactive protein concentration of more than 20 mg/L; had chronic fever; had a psychiatric disorder precluding understanding of information on trial-related topics; were prescribed concurrent treatment with other experimental drugs or treatment in another clinical trial within 30 days before trial entry; had any serious underlying medical condition that could impair their ability to participate in the trial; or if they had a known allergy or hypersensitivity to any of the study drugs. Participants were tested for malaria by a rapid diagnostic test and microscopy, and for helminthic infections by stool ova and parasite examination; those testing positive were treated according to national guidelines and were eligible for enrolment.
Randomisation and masking
Participants were enrolled and randomly assigned by study personnel. Participants were randomly assigned in a 1:1 ratio to receive either intravenous ferric carboxymaltose or oral iron, stratified by haemoglobin concentration (<70 g/L, 70–100 g/L, or >100 g/L) and site. The trial statistician generated the treatment allocation scheme in advance by computer. Opaque envelopes were prepared on site according to this scheme by people who were independent from the trial, and were sealed and marked on the outside with a sequential number and the stratification information. The randomisation processes were subject to monitoring and verification by the trial statistician; it was found that the haemoglobin stratification was not properly applied during the first 68 random assignments, resulting in simple randomisation being applied for those enrolments. The process was corrected and continued as planned. Importantly, allocation concealment was maintained throughout the trial.
Treatment was started within 14 days of screening and 7 days of childbirth. Eligible participants who did not deliver within 14 days of screening could be rescreened on a predefined subset of parameters including haemoglobin and ferritin concentrations.
Ferric carboxymaltose (Vifor Pharma, Villars-sur-Glâne, Switzerland) was administered intravenously according to the manufacturer at a dose determined by the haemoglobin concentration and bodyweight. Participants with a bodyweight of 35 kg to less than 70 kg and a haemoglobin concentration of 100 g/L or more received 1000 mg ferric carboxymaltose in one dose; participants with a bodyweight of 35 kg to less than 70 kg and a haemoglobin concentration of less than 100 g/L, or a bodyweight of 70 kg or more and a haemoglobin concentration of 100 g/L or more, received 1500 mg in two doses at least 7 days apart; participants with a bodyweight of 70 kg or more and a haemoglobin concentration of less than 100 g/L received 2000 mg in two doses at least 7 days apart.
Oral iron treatment consisted of three dried ferrous sulphate tablets of 200 mg containing 60 mg of elementary iron and 5 mg of folic acid every morning. Participants were advised by a trained nurse about intake modalities (30 min before food, tea, or coffee; or in case of side-effects, with a meal or in two separate doses per day [two tablets in the morning and one in the evening]). Oral treatment was to be taken for 3 months after correction of anaemia, defined as a haemoglobin concentration of more than 115 g/L. Drug accountability logs were used to record the number of tablets issued at each visit and pill counting was done to monitor adherence.
At enrolment and all follow-up visits, whole blood cell counts were analysed on a Sysmex XE-2100 five populations analyser (Sysmex Europe, Norderstedt, Germany). Ferritin, hepatic function parameters, serum creatinine, and C-reactive protein were measured on a COBAS INTEGRA 400 plus analyser (Roche Diagnostics, Mannheim, Germany). Data from laboratory analyses were extracted from the analysers and linked to the respective clinical data by a common unique identifier.
The primary outcome was the proportion of participants with a normalised haemoglobin concentration (>115 g/L) at 6 weeks after treatment initiation. Secondary outcomes were the proportion of participants with corrected iron deficiency (defined as a normalised serum ferritin concentration of >100 μg/L) at 6 weeks after treatment initiation, haemoglobin and ferritin best responses (highest values), time to haemoglobin and ferritin best responses, adherence to study medication, adverse events, and wellbeing, as measured by the SF-36.
Sensitivity, specificity, and the negative and positive predictive values of erythrocyte indices for the diagnosis of iron deficient anaemia in Tanzania will be reported later.
We assumed that 85% of participants in the intravenous iron group and 70% in the oral iron group would have a normalised haemoglobin concentration at 6 weeks. To detect a difference in response rates of 15%, with a one-sided type I error probability of 0·05 and 80% power, 95 participants per group were required (assuming a two-sided type I error of 0·05, this sample size yields 70% power). An additional 20% was added to the sample size to allow for loss to follow-up, yielding 230 participants overall.
In preplanned sensitivity analyses, we adjusted for covariates that appeared unbalanced between groups at baseline by visual inspection or were associated with differential follow-up. We did the following post-hoc analyses: sensitivity analyses in which participants with missing primary outcome data were classified as not having haemoglobin normalisation at 6 weeks; sensitivity analyses excluding the first 68 participants for whom the randomisation stratification was not applied; unadjusted analyses; and analyses considering thresholds of more than 120 g/L for haemoglobin
and more than 30 μg/L for ferritin. We assessed a-priori effect modification by baseline haemoglobin concentration. Haemoglobin and ferritin best responses and times to best response were analysed descriptively.
For the intravenous iron group, adherence was assessed by the number and proportion of participants receiving infusions according to the protocol. There were some instances of incorrect dose calculation, for example screening values of haemoglobin being used to determine dosage, or participants with bodyweight, haemoglobin value, or both on the border of the thresholds for dosing calculations (that is, bodyweight equal to 70 kg or haemoglobin concentration of 100 g/L) mistakenly allocated to the wrong dosage. We therefore also defined adequate dosage, permitting such participants to be allocated to doses defined by the adjacent bodyweight or haemoglobin values, and considering those who received at least their protocol-defined dose as adherent. For the oral iron group, adherence was assessed by pill count at every visit.
Analyses were done using Stata (version 15). A data safety monitoring board reviewed the trial progress and data at a meeting in September, 2016, and with a further short report in November, 2016. The trial is registered at ClinicalTrials.gov, NCT02541708.
Role of the funding source
The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. All authors had full access to all the data in the study and had final responsibility for the decision to submit for publication.
Table 1Baseline characteristics
Data are n (%) or median (IQR).
Table 2Adherence to intravenous and oral iron
Data are median (IQR) or n/N (%).
Quality of health scores on the SF-36 improved following childbirth in both groups, and remained close to 100% for most of the health domains (appendix p 18).
In this trial, intravenous iron substitution with ferric carboxymaltose was associated with significantly faster and more durable normalisation of haemoglobin and ferritin concentrations, compared with oral iron substitution. Overall, 80% of participants in the intravenous iron group with primary outcome data had normalised haemoglobin concentrations after 6 weeks, compared with only 51% of participants in the oral iron group. On average, ferritin concentrations remained below normal among those in the oral iron group, whereas 80% of participants in the intravenous iron group had a normalised ferritin concentration after 1 year, suggesting a restoration of iron stores. This might be an important gain in health for mother and child, particularly in subsequent pregnancies because many women might not receive medical care between births. In prespecified analyses, there was a numerically greater benefit of intravenous iron in terms of haemoglobin normalisation among participants with a lower baseline haemoglobin concentration at treatment initiation (<100 g/L vs ≥100 g/L), suggesting this population might be an important target group with greater benefit, but the p value was large.
Intravenous iron substitution with ferric carboxymaltose is routinely done in high-income settings. Our study shows that intravenous iron substitution with ferric carboxymaltose can be delivered safely in district hospitals in a resource-limited setting. Ferric carboxymaltose can be stored at room temperature (<30°C) and has a shelf life of 3 years from manufacture. With ferric carboxymaltose, the total required dose can be delivered in just one or two doses, and without a test dose as is needed for iron sucrose. Hospital attendance for delivery is used to provide medical prevention strategies such as vaccinations, and provides an ideal opportunity to also deliver a straightforward and effective iron supplementation strategy before discharge, after which individuals might become lost to medical care.
However, we did not observe any grade 3 or grade 4 adverse events in the intravenous iron group. The safety concerns for anaphylactic reactions frequently observed with the application of iron dextran are not observed with iron carboxymaltose, neither in high-income settings,
nor in our study in a resource-limited setting. However, although iron dextran is registered for use in many low-income countries, including Tanzania, ferric carboxymaltose is not yet.
Despite the beneficial effects of intravenous iron on haemoglobin and ferritin normalisation, we did not observe any differences in quality-of-life between the intravenous and oral iron groups. This might be due to a ceiling effect, with many of the quality of life scores close to 100%, a limitation that has been noted by others.
and the risks associated with transfusions are high.
Fourth, we did not directly analyse acceptability in detail, but only a small proportion of eligible individuals explicitly refused consent. Furthermore, almost all participants in the intravenous iron group received at least one dose, and the vast majority received an adequate dose. Thus, our study shows the feasibility and acceptability of intravenous iron in this population. By contrast, long-term adherence to oral iron was poor, reflected in the non-repletion of iron stores, thus limiting the potential of oral iron for correction of iron deficiency anaemia. Last, intravenous administration of iron preparations can cause hypophosphataemia, which is mainly transient and clinically irrelevant. According to published studies, there are only isolated cases of hypophosphataemia requiring treatment, mainly in patients with existing risk factors and after prolonged exposure to high-dose intravenous iron.
Measuring of serum phosphate was not included in the protocol of the study, which precluded collection of additional information on this topic.
In conclusion, individuals with iron deficiency anaemia at delivery had a better haemoglobin response and more complete repletion of iron stores over time with ferric carboxymaltose intravenous iron than with oral iron substitution. This study adds important data on the benefits of ferric carboxymaltose and provides evidence for its benefits in a low-income setting, thus paving the way for approval of this drug in such settings. Our data can now be discussed and shared with international funding organisations, ministries of health, and local health-care providers to inform how to best transfer this treatment to sub-Saharan countries and which financial mechanisms might be best suited to support it. Broad use in a peripartum setting might help reduce iron deficiency anaemia where there is the greatest burden. This study was done following childbirth in individuals with iron deficiency anaemia in Tanzania and our results are in line with those observed in similar populations in high-income settings. Therefore, one can anticipate similar benefits in other patient populations with iron deficient states in resource-limited settings as already observed in high-income countries, such as individuals with chronic kidney diseases, inflammation, or heart insufficiency. Studies in these patient populations in low-income countries are needed, especially because such populations are increasing. Further studies are needed to confirm the safety and cost-effectiveness of ferric carboxymaltose in other low-income settings with different socioeconomic and health-care contexts, and to assess the potentially long-term benefits of a highly effective iron substitution on mother and child health.
MT and SM-M conceived and designed the study, and oversaw its implementation, analysis, and write-up. FV and TRG planned the statistical analyses. FV generated the random allocation and did the statistical analyses. OL led the field implementation of the study and was responsible for data entry. KDM, PA, and AI contributed to the field implementation of the study and did data entry. BS supervised storage and distribution of medication and supervised pill counting. SM and CD did the laboratory analyses. AK and SS implemented the quality assurance plan. FV, AK, and SS led the data management. SA, CD, MT, and SM-M provided trial oversight. FV, AK, and SM-M wrote the first draft of the manuscript. Data were accessible to FV, AK, and SM-M after data freezing on Aug 29, 2018; FV, AK, and SM-M checked the data. All authors read and approved the manuscript.
Declaration of interests
We declare no competing interests.
We thank the women who participated in the study, the staff of the participating hospitals, and the staff of Ifakara Health Institute. We thank Jörg Halter, Abel Makubi, and Amanda Ross for their work as members of the Data Safety Monitoring Board. We thank the regional monitors Rose Minja and Beatrice N Nyakundi, and nurse assistants Janeth Kiwasila and Lydia Balehetse. Ferric carboxymaltose was donated by Vifor Pharma (Villars-sur-Glâne, Switzerland). Sysmex analyser was donated by Sysmex Europe (Norderstedt, Germany). This study was funded by Vifor Pharma (unrestricted educational grant), R Geigy-Stiftung, Freiwillige Akademische Gesellschaft, and Swiss Tropical and Public Health Institute.
Global, regional, and national disability-adjusted life-years (DALYs) for 333 diseases and injuries and healthy life expectancy (HALE) for 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016.
Lancet. 2017; 390: 1260-1344
Worldwide prevalence of anaemia 1993–2005: WHO global database on anaemia.
World Health Organization, Centers for Disease Control and Prevention Atlanta,
Anaemia in low-income and middle-income countries.
Lancet. 2011; 378: 2123-2135
Severe acquired anaemia in Africa: new concepts.
Br J Haematol. 2011; 154: 690-695
Maternal iron deficiency anemia affects postpartum emotions and cognition.
J Nutr. 2005; 135: 267-272
Guidelines for the use of iron supplements to prevent and treat iron deficiency anemia.
International Nutritional Anemia Consultative Group (INACG), World Health Organization,
Poor economics: a radical rethinking of the way to fight global poverty.
New York, NY2011
Ferrous sulfate supplementation causes significant gastrointestinal side-effects in adults: a systematic review and meta-analysis.
PLoS One. 2015; 10e0117383
Meta-analysis of efficacy and safety of intravenous ferric carboxymaltose (Ferinject) from clinical trial reports and published trial data.
BMC Blood Disord. 2011; 11: 4
Large-dose intravenous ferric carboxymaltose injection for iron deficiency anemia in heavy uterine bleeding: a randomized, controlled trial.
Transfusion. 2009; 49: 2719-2728
Intravenous ferric carboxymaltose compared with oral iron in the treatment of postpartum anemia: a randomized controlled trial.
Obstet Gynecol. 2007; 110: 267-278
Evaluation of the reported rates of severe hypersensitivity reactions associated with ferric carboxymaltose and iron (iii) isomaltoside 1000 in Europe based on data from EudraVigilance and VigiBase™ between 2014 and 2017.
Drug Saf. 2019; 42: 463-471
Comparative risk of anaphylactic reactions associated with intravenous iron products.
JAMA. 2015; 314: 2062-2068
Direct comparison of the safety and efficacy of ferric carboxymaltose versus iron dextran in patients with iron deficiency anemia.
Anemia. 2013; 2013169107
The available intravenous iron formulations: history, efficacy, and toxicology.
Hemodial Int. 2017; 21: S83-S92
Differences in spontaneously reported hypersensitivity and serious adverse events for intravenous iron preparations: comparison of Europe and North America.
Arzneimittelforschung. 2011; 61: 267-275
Hypersensitivity reactions and deaths associated with intravenous iron preparations.
Nephrol Dial Transplant. 2005; 20: 1443-1449
Immunol Today. 1982; 3: 132-138
SF-36 health survey update.
Spine. 2000; 25: 3130-3139
How to score version 2 of the SF-36 health survey.
Relative risks and confidence intervals were easily computed indirectly from multivariable logistic regression.
J Clin Epidemiol. 2007; 60: 874-882
Haemoglobin concentrations for the diagnosis of anaemia and assessment of severity.
Maternal and fetal risk factors for stillbirth in northern Tanzania: a registry-based retrospective cohort study.
PLoS One. 2017; 12e0182250
Control of iron deficiency anemia in low- and middle-income countries.
Blood. 2013; 121: 2607-2617
Validation of the Kiswahili version of the SF-36 health survey in a representative sample of an urban population in Tanzania.
Qual Life Res. 1999; 8: 111-120
A Kiswahili version of the SF-36 health survey for use in Tanzania: translation and tests of scaling assumptions.
Qual Life Res. 1999; 8: 101-110
The global need and availability of blood products: a modelling study.
Lancet Haematol. 2019; 6: e606-e615
High prevalence and poor linkage to care of transfusion-transmitted infections among blood donors in Dar-es-Salaam, Tanzania.
J Viral Hepat. 2019; 26: 750-756
A pooled analysis of serum phosphate measurements and potential hypophosphataemia events in 45 interventional trials with ferric carboxymaltose.
European Crohn’s and Colitis Organisation (ECCO),
Vienna, AustriaFeb 12–15, 2020
Published: November 24, 2020
© 2020 The Author(s). Published by Elsevier Ltd.