Intravenous versus oral iron for iron deficiency anaemia in pregnant Nigerian women (IVON): study protocol for a randomised hybrid effectiveness-implementation trial

Background: Anaemia in pregnancy is highly prevalent in African countries. High-dose oral iron is the current recommended treatment for pregnancy-related iron deficiency anaemia (IDA) in Nigeria and other African countries. This oral regimen is often poorly tolerated and has several side effects. Parenteral iron preparations are now available for the treatment of IDA in pregnancy but not widely used in Africa. The IVON trial is investigating the comparative effectiveness and safety of intravenous ferric carboxymaltose versus oral ferrous sulphate standard-of-care for pregnancy-related IDA in Nigeria. We will also measure the implementation outcomes of acceptability, feasibility, fidelity, and cost-effectiveness for intravenous ferric carboxymaltose. Methods: This is an open-label randomised controlled trial with a hybrid type 1 effectiveness-implementation design, conducted at 10 health facilities in Kano (Northern) and Lagos (Southern) states in Nigeria. A total of 1056 pregnant women at 20–32 weeks’ gestational age with moderate or severe anaemia (Hb < 10g/dl) will be randomised 1:1 into two groups. The interventional treatment is one 1000-mg dose of intravenous ferric carboxymaltose at enrolment; the control treatment is thrice daily oral ferrous sulphate (195 mg elemental iron daily), from enrolment till 6 weeks postpartum. Primary outcome measures are (1) the prevalence of maternal anaemia at 36 weeks and (2) infant preterm birth (<37 weeks’ gestation) and will be analysed by intention-to-treat. Maternal full blood count and iron panel will be assayed at 4 weeks post-enrolment, 36 weeks’ gestation, delivery, and 6 weeks postpartum. Implementation outcomes of acceptability, feasibility, fidelity, and cost will be assessed with structured questionnaires, key informant interviews, and focus group discussions. Discussion: The IVON trial could provide both effectiveness and implementation evidence to guide policy for integration and uptake of intravenous iron for treating anaemia in pregnancy in Nigeria and similar resource-limited, high-burden settings. If found effective, further studies exploring different intravenous iron doses are planned. Trial registration: ISRCTN registry ISRCTN63484804. Registered on 10 December 2020 Clinicaltrials.govNCT04976179. Registered on 26 July 2021 The current protocol version is version 2.1 (080/080/2021). The IVON trial will adopt a hybrid type 1 effectiveness-implementation design, where the effect of an intervention is evaluated while documenting observations about its implementation [30]. The effectiveness study is a two-arm, open-label, individually randomised controlled trial (RCT) that assesses the comparative effectiveness and safety of FCM versus oral iron in treating IDA among pregnant Nigerian women. Intervention effectiveness will be assessed through analyses for the primary and secondary measures, while implementation will be documented through reporting on the implementation outcomes of (1) acceptability, (2) feasibility, (3) fidelity, and (4) cost-effectiveness (Fig. ​(Fig.11). IVON trial effectiveness-implementation design We used the SPIRIT checklist for protocol reporting [31] to guide the development of our protocol (Additional file 1). The study will be conducted at health facilities in Kano and Lagos states, representing the North-West and South-West zones, respectively, which, by population, are the two largest states in Nigeria. Rates of antenatal care utilisation are 63.4% and 51.4% in urban and rural areas of Kano, respectively [32], and 81.4% in Lagos [33]; estimates for delivery by skilled personnel are 21.5% and 83.6% in Kano and Lagos, respectively [34]. The prevalence of AIP in Kano and Lagos is reported at 17.5 to 75% [35, 36] and 35.3 to 87.2% [37, 38], respectively. Five study sites (one tertiary, two secondary, and two primary health facilities) were selected per state, for a total of 10 sites (Fig. ​(Fig.2).2). Site eligibility criteria were exclusively publicly funded primary, secondary, or tertiary health facility; antenatal clinic attendance of at least 60 pregnant women per month; delivery rate of at least 20 women per month; consistent 24-h routine vaginal delivery services; and onsite testing for Hb, HIV, and malaria. IVON clinical trial flowchart Pregnant women with: The intervention drug under investigation is intravenous ferric carboxymaltose (Ferinject®, FCM), and the standard-of-care drug is oral ferrous sulphate (Fesulf®). FCM is given as a single dose of 20 mg/kg up to a maximum of 1000 mg diluted in 200 ml of 0.9% sodium chloride and infused over 15–20 min. FS will be given as one 200-mg tablet containing 65 mg of elemental iron, three times daily, to be taken 1 h before meals or 2 h after meals until 6 weeks postpartum. Women in both intervention and control groups will also receive 5 mg of folic acid and 100 mg of vitamin C daily throughout pregnancy. In addition, participants in both groups will receive daily reminders via short message service (SMS) sent directly to their mobile phone numbers, to take their oral medications. Pill counts will also be done at each study visit. Primary outcomes are the prevalence of maternal anaemia (Hb <10 g/dl) at 36 weeks’ gestation and the incidence of preterm birth (<37 weeks’ gestation) (Fig. ​(Fig.1).1). Secondary maternal outcomes include an increase in maternal haemoglobin concentration 4 weeks after administration; the prevalence of maternal depression at 36 weeks’ gestation and at 2 weeks postpartum as assessed with the Edinburgh Postnatal Depression Scale (EPDS); safety of trial drugs as determined by the prevalence of severe adverse events; incidence of severe maternal events such as shock and sepsis as determined by physician diagnosis or medical record documentation; incidence of postpartum haemorrhage determined by visually estimated blood loss >500 ml for vaginal delivery and >1000 ml for caesarean section; and need for blood transfusion. Secondary infant outcomes include incidence of low birthweight (<2.5 kg); prevalence of stillbirth; small-for-gestational age (birthweight below the 10th percentile for gestational age), neonatal death (within 28 days of birth); and proportion of infants being breastfed at 1, 2, and 4 weeks of life, and having received Bacille–Calmette–Guerin, oral polio, and hepatitis vaccinations within 4 weeks of life (Table ​(Table22). Specifically defined study outcomes A score of ≥10 on the Edinburg Postnatal Depression scale (EPDS) will be considered indicative of postpartum depression, as this has been shown to be highly sensitive for diagnosing minor depression (Table ​(Table2)2) [39]. All women screened positive for depression will be referred to the collaborating psychiatrist for further assessment. Implementation outcomes to be assessed are acceptability (the state of meeting stakeholder needs, preferences, or expectations), feasibility (the state or degree of the intervention being easily or conveniently done), and fidelity (the degree to which the intervention is delivered as intended). In addition, costs of the intervention and control treatment options to pregnant women and to the health system will be collected (Table ​(Table22). Participants will be enrolled at 20–32 weeks’ gestational age and followed up until 6 weeks postpartum (Fig. ​(Fig.3).3). Pregnant women presenting to antenatal care clinics at study sites will be approached by research nurses and counselled on anaemia in pregnancy and the IVON trial. Consenting pregnant women will be screened for Hb by finger pinprick, using the Hemocue® haemoglobinometer. Women with Hb < 10g/dl who meet other eligibility criteria will be further counselled about the study, including risks, benefits, and what is expected as a participant, as part of the informed consent process. Potential participants will be able to sign informed consent forms immediately or to sign later if they wish to discuss with male partners or family members. Spirit figure showing specific time points in IVON trials. Note: denotes variables that will be assessed at each visit from the selected time point (X) to the end point depicted by the double-headed arrow. Acceptability of intervention will be assessed pre-trial, and findings used in protocol modification Site research nurses and coordinators will all be required to maintain Good Clinical Practice certification and will be trained on the research protocol. Recruitment is targeted for completion by end of July 2023. Per the multi-country FER-ASAP (FERric carboxymaltose-Assessment of SAfety and efficacy in Pregnancy) study in Europe, Asia, and Australia [40], the expected levels of anaemia correction for oral iron and FCM in pregnancy are approximately 70% and 84%, respectively. At 5% significance, 90% power and adjusting for 15% attrition [41], 1056 pregnant women (528 in each study arm) was the estimated sample size to detect a 14% difference in the correction of AIP between the control (70%) and intervention groups (84%). Eligible participants will undergo individual randomisation and allocation. Randomisation and concealment will be done by Sealed Envelope, UK (www.sealedenvelope.com), through a web-based randomisation service, that will randomise in a 1:1 ratio in blocks stratified according to the study site. Once consent is signed, the research nurse enters enrolment details such as biodata and obstetric history of the participant into an electronic handheld tablet. On submission of the electronic form, the Sealed Envelope software will randomly allocate the participant to intervention or control. There is no blinding except during the data analysis. Given the obvious morphological differences of the intervention (intravenous solution) and control (oral tablet) drugs, both participants and study staff will be aware of study arm allocations following randomisation. Statisticians will be blinded to treatment assignment during analysis. For data on effectiveness outcomes, following randomisation, data are collected electronically, using case report forms designed with REDcap electronic data capture tools, hosted at the College of Medicine, University of Lagos [42, 43]. General and obstetric physical examinations are performed at baseline and at each study visit from enrolment to 6 weeks postpartum. Hb and full blood count measurements are done for all study visits and for five of six visits, respectively. Malaria is assessed with a rapid diagnostic test (SD BIOLINE Malaria Ag P.f) at enrolment and whenever participants have symptoms consistent with malaria. All participants diagnosed with malaria are treated with Arthemeter-Lumefantrine as per national guidelines [44]. Serum ferritin alone is done at enrolment, but complete iron profile comprising serum ferritin, serum iron, total iron binding capacity, and transferrin saturation are done at 4 weeks after initiation of the study drug, at 36 weeks, at delivery, and at 6 weeks postpartum. Maternal serum phosphate level is assayed at enrolment, 4 weeks after iron infusion/oral iron commencement, on the day of delivery, and at 6 weeks postpartum. Cord blood phosphate is assayed on the day of delivery. Participants are seen in a routine antenatal clinic every 4 weeks till 28 weeks’ gestation, and every 2 weeks until 36 weeks, then weekly until delivery. Thereafter, they are seen at 2 and 6 weeks postpartum (Fig. ​(Fig.33). To minimise loss-to-follow up in the RCT, retention counselling for study participants will be extended to male partners and relevant family members during antenatal visits and will emphasize facility delivery at study sites. Also, verification of locator information (phone numbers and residential addresses) will be done at every clinic visit. Finally, research nurses will send SMS appointment reminders to participants, who will also be tracked in the community (at their homes or other places of delivery, if applicable), when they do not show up for expected study visits. For data on implementation outcomes, besides the CFIR-guided qualitative data collected during the formative study, quantitative data on acceptability and feasibility was collected from among healthcare workers, using the Acceptability of Intervention Measure (AIM) and Feasibility of Intervention Measure (FIM) tools developed by Weiner et al. [45]. For the trial proper, concurrent qualitative and quantitative assessments of acceptability and feasibility will be repeated at the end of the study. Exit interviews will be conducted quantitatively through surveys among 25% randomly selected participants in each arm of the trial and qualitatively through in-depth interviews using semi-structured interview guides among purposively selected participants (Table ​(Table33). Timeline of data collection for implementation outcomes AIM acceptability of intervention measure, FIM feasibility of intervention measure, FGD focus group discussions, KII key informant interview For cost data needed for the cost-effectiveness analysis, direct costs such as cost of medications including the intravenous FCM and oral FS, folic acid, and vitamin C, out-of-pocket expenses by users including transportation to care, opportunity costs of antenatal visits, short admission stay for infusion of intravenous FCM and for management of any complications and health system costs such as laboratory consumables, supplies, and diagnostics will be collected. Patient cost data will be collected from IVON trial participants at follow-up visits during the antenatal and postpartum periods. Health system costs will be derived from hospital administrators and from data available in published studies from Nigeria. Data collection and reporting for both effectiveness and implementation outcomes will be guided by standard operating procedures that will be periodically updated, shared with, and communicated to the study staff. Effectiveness data will be captured in electronic case report forms designed for each study visit, using assigned codes, and uploaded in real time to the central REDcap server after verification by each site coordinator. Implementation data will be captured in standard study tools or as transcripts in the study database, which is password protected. For confidentiality and privacy, all qualitative and quantitative participant data will be de-identified from the point of data collection to curation and publication, and only the principal investigator, project manager, and data manager will have access to the study data. An interim analysis will be conducted by the Data and Safety Monitoring Committee after 50% of study participants have been recruited and fully followed up, with the objective to (1) evaluate the overall safety of the interventional intravenous iron; (2) examine the adequacy of assumptions made for sample size estimation; (3) decide, based on clear evidence of efficacy or lack thereof, whether early termination of the trial is warranted; and (4) determine whether there are unanticipated reasons for early trial termination or modification. There will be considerations for study discontinuation if there is a ≥28% difference (double the 14% assumed difference) in the prevalence of women with improved anaemia in the intervention group compared to the control group. Categorical variables will be expressed as frequencies and percentages. For continuous variables, a Shapiro–Wilk test of normality will be performed, and normally distributed data will be presented as means with standard deviations, while non-normally distributed data will be presented as medians with interquartile ranges. The main analysis will be by intention-to-treat. The intention-to-treat (ITT) population will include all subjects who were enrolled and randomised. Baseline characteristics will be compared between the intervention and control groups using Pearson’s chi-square (for categorical variables), Student’s t-test (for normally distributed continuous variables), or Mann–Whitney U (for non-normally distributed continuous variables). Paired t-test or Wilcoxon signed rank t-test will be used to assess the intra-group change in Hb, ferritin, and transferrin saturation concentrations from baseline to 6 weeks postpartum. Univariable and multivariable binary logistic regression will be used to compare the proportions of women achieving anaemia correction, with the assigned study group as the primary explanatory variable. Simple and multiple linear regression modelling will assess changes in continuous variables. Kaplan–Meir plots and unstratified log-rank test will be used to explore time-to-anaemia correction. The above analysis of the proportions achieving anaemia correction and time-to-anaemia correction will be repeated in subgroups of baseline iron deficiency (versus no iron deficiency) and iron deficiency anaemia (versus no iron deficiency anaemia). Effect modification of the effectiveness of interventions by baseline iron deficiency and iron deficiency anaemia will be evaluated using the likelihood ratio tests and p-values reported. Iron deficiency will be defined in accordance with the World Health Organization guideline as serum ferritin < 15μg/l (ref), and iron deficiency anaemia will be defined as iron deficiency in the presence of anaemia. Sensitivity analysis will be conducted in the modified intention-to-treat population. The modified intention-to-treat population will include participants who complete 36 weeks of delivery and are eligible for assessment of the primary endpoint. Two-tailed test of hypothesis and 95% confidence interval is the assumed statistical level of significance. Statistical analysis will be conducted using STATA SE version 16.0 (STATA Corp, TX, USA) software. All qualitative interviews will be recorded and transcribed verbatim. Transcribed data will be coded both deductively (informed by CFIR) [46] and inductively, to generate further themes from the data. Content analysis will then be used to capture underlying ideas and patterns. AIM and FIM survey responses will be interpreted per Weiner et al. guidance [45]. Fidelity will be calculated as a proportion of IV iron administrations performed that met the minimum set of checklist items for each FCM administration completed. The analysis will be done using descriptive statistics and test of difference regressions. Measures of cost and effectiveness will be obtained using primary client-level data on cost, hospital costs for providing care, and results of clinical outcomes from the effectiveness evaluation. Specifically, cost elements of both oral and intravenous iron treatment options will be quantified and comparatively assessed at an individual client level, using a decision tree [47, 48]. Comparative analyses of direct medical costs of intravenous FCM and oral ferrous sulphate costs and costs for administration will be included. Analysis of all costs borne by clients from potential adverse events, and from any loss of productivity while incapacitated by these adverse events, will be conducted for all affected participants. Empirical evidence on probabilities will be obtained from the trial as well as from secondary data sources. Analysis will be conducted from a limited societal perspective [49]. The incremental cost-effectiveness ratio (ICER) will be obtained from mean values for costs and effectiveness, the latter defined by significant differences in the prevalence of anaemia at 36 weeks, and in the incidence of preterm birth. The confidence interval for the ICER will be derived using non-parametric bootstrapping to quantify sampling uncertainty. Uncertainty will be addressed by a univariate sensitivity analysis, which aims to identify parameters whose uncertainty has the largest influence on the ICER. In addition, a probabilistic sensitivity analysis, in which all parameters with uncertainty will be varied within their confidence intervals, will provide a credible time interval for the estimated result of the ICER. No discounting will be incorporated in the analysis as the analytic horizon is less than a year. IVON has an Administrative Core, a Steering Committee, Clinical Trial Monitors, and a Data and Safety Monitoring Committee (DSMC). The Administrative Core comprises the principal investigator and a full-time project manager who provide administrative oversight, manage research activities, and coordinate all internal (implementing staff) and external (collaborator) meetings. The Steering Committee includes the principal investigator, all co-investigators, and site coordinators; provides cross-functional leadership and direction and project governance; and ensures adherence to the research protocol to facilitate the achievement of project goals. Clinical Trial Monitors will conduct weekly site monitoring visits to ensure and verify human subject protection and conduct documentation audits and verification of reported trial data. The DSMC is an independent, five-person external monitoring committee including an obstetrician, a haematologist, a statistician, an experienced trial pharmacist, and a paediatrician with research expertise. The DSMC met once before the trial started and will meet every 6 months during the study. The DSMC will meet after the 50th participant has completed the study to assess the progress of the RCT, recruitment rate, and effect of drugs and decide whether to modify, terminate, or continue the clinical trial. In our study, an adverse event (AE) will be defined as any untoward medical occurrence in a participant regardless of the possibility of a causal relationship to the trial drug. All adverse events that occur from the point of enrolment and throughout the study duration will be collected non-systematically from each participant. Participants will be routinely asked at every study visit “do you have any symptoms now or noticed any since your last visit?” If a participant reports an AE at enrolment but before the administration of study intervention, the AE will be reported as not related to the trial drug. All adverse events reported after enrolment into the study and until the end of the study will be defined and graded for severity using the US National Cancer Institute Common Terminology Criteria for Adverse Events version 5.0 [50] and documented as such for analysis. The suspected relationship to trial drugs will be determined by the principal investigator and documented along with the interventions given, and the outcome. A serious adverse event (SAE) is any event which is fatal or life-threatening, or results in persistent or significant disability/incapacity, or constitutes a congenital anomaly/birth defect, or requires inpatient hospitalization or prolongation of existing hospitalization. Other criteria include an event that is sufficiently significant as to require medical or surgical intervention to prevent one of these outcomes or is a significant hazard as determined by the Data Safety Monitoring Board. An adverse event that meets the criteria for a SAE between enrolment and end of the study will be reported to the local Institutional Review Board (IRB) as an SAE. Discontinuation of the study drug may be considered if a participant experiences a severe adverse drug event of grade 3 or higher [50]. Decision for discontinuation will be made by the principal investigator and/or the study participant. Study personnel will document the circumstances and data leading to the discontinuation. Investigators will also determine whether an AE is a suspected unexpected serious adverse reaction (SUSAR) given the participants’ clinical course, previous medical conditions, and concomitant medications. A SUSAR is defined as an untoward and unintended response to a trial drug, which is not listed in the product information, and meets one of the following serious criteria: results in death, is life-threatening, requires hospitalization or prolongation of an existing hospitalization, results in persistent or significant disability or incapacity, or is a congenital anomaly or birth defect. SUSAR report will be expeditiously submitted to health regulatory authorities. The costs of managing all adverse drug events related to trial drugs will be borne by the study. Using an intention-to-treat analysis, the absolute risk, the frequency/incident rate, and severity grade of anticipated adverse drug reactions such as abdominal pain, nausea and vomiting, hypotension, and hypophosphatemia for participants in each study arm, the incident rate of any serious adverse reactions (SAR) and suspected unexpected serious adverse reaction (SUSAR) for each arm of the study will be determined. Serious adverse reactions (SARs) are SAEs that are thought to be causally linked to the trial drug. The mean and SDs or median and interquartile range for serum phosphate level will be calculated as appropriate, while extreme values will be stated if observed.