The main objective of this review was to identify gaps in the evidence for the prevention and treatment of maternal anaemia in pregnancy and up to 1 year postpartum, and to inform the research agenda. This topic remains a concern to health care professionals and public health specialists, as well as to women themselves, but the literature addressing this field is widely dispersed over a number of specialist publications/fields and has not been previously systematically collated.
One of the first striking gaps in the evidence base is in the management of maternal anaemia up to one year postpartum (Table 1). Secondly, across all identified systematic reviews, 25 included interventions for the prevention of anaemia, compared with only six reviews that included interventions for the treatment of anaemia. Thirdly, none of the reviews reported on the best methods of screening for anaemia. Finally, the majority of antenatal reviews evaluated the role of nutritional supplementation, including the use of iron which was consistently shown to improve haematological parameters and decrease the risk of anaemia at term. It was, however, impossible to make definitive statements about optimum dose or regimen, due to the heterogeneity between trials.
There was limited evidence to demonstrate that riboflavin, folic acid, vitamin A and the use of erythropoietin could enhance a response to iron. It is, therefore, unclear how or whether these other supplements should be used in current practice. No reviews examined the use of dietary modification or food support for the management of anaemia, and no reviews reported on cost-effectiveness of different approaches.
No reviews reported on clinical symptoms of anaemia and few reported on the correlation between laboratory and clinical outcomes. Despite evidence of improvement in haematological parameters with use of iron there was limited evidence to indicate that the reduction in antenatal maternal anaemia had any clinical impact for mother or infant, other than reducing the need for blood transfusion. There was some evidence to suggest that prompt treatment of postnatal anaemia may be beneficial in preventing postnatal depression and improving lactation, infant development and maternal bonding but the quality of this review was poor therefore results should be interpreted with caution. Data on adverse effects such as maternal exhaustion were lacking from the systematic review literature. A number of reviews reported on the use of different antenatal care programs for the prevention of antenatal anaemia, but there was no evidence that any type of program, including frequency of visits, was superior.
Recommendations do exist for best practice regarding the investigation and management of anaemia but these guidelines also seem to reflect many of the evidence uncertainties highlighted in this overview. The National Institute for Clinical Excellence (NICE) provides guidance on the management of antenatal care  and postnatal care , but in the postnatal guidelines there is no mention of the word ‘anaemia’. The RCOG green-top guideline for blood transfusion in obstetrics  highlights the importance of screening and treating anaemia, with oral or parental iron, in order to reduce the chances of later blood transfusion. Yet within these guidelines it is indicated that these are only ‘good practice points’ indicating that high-quality, systematically derived evidence is lacking.
Methodological quality of the included systematic reviews was variable with less than half assessed as being of high quality, thus consequently many of these systematic reviews are of limited utility for those seeking an evidence base to practice. 11 of the 12 high quality reviews fell across prevention of anaemia in the antenatal period and thus the findings and recommendations for practice from these reviews would be deemed to be more robust. Only one review of high quality was found for the postnatal period. Further systematic reviews in the postnatal period should meet the criteria for quality appraisal (PRISMA ) and provide information both on existing systematic reviews and the novel contribution of the new review.
A limitation of this review is the variation in practices and populations within and between countries, which may impact on the generaliseability of findings, although we partially addressed this by excluding reviews focussing on resource poor countries. A limitation of the methods used for this review of reviews is that, inevitably, there may be some recently published trials that have not yet been included in systematic reviews and thus are not included in our review. There is also some overlap in analyses of the same primary RCTs between systematic reviews; we have clearly identified where this has occurred.
The findings from this review of reviews have implications for new primary research and the research agenda. Anaemia management remains an uncertain area of research focus, despite its clinical importance. The identification or management of maternal anaemia was the main stated aim in only seven reviews [14–16, 27, 28, 33, 34] and overall, the number of RCTs that specifically reported on outcomes relevant to maternal anaemia was small (median 7, interquartile range 1–23).
A further priority is to develop standardised approaches to the reporting of laboratory outcomes as the heterogeneity of populations and settings, drug dosages and regimens and definitions of anaemia precluded quantitative analysis for many reviews. Outcomes relevant to the experiences and wellbeing of pregnant women and new mothers need to be included in future randomised controlled trials. Additional work is needed to identify the most effective way of managing anaemia in the postnatal period; validating an effective screening policy for both the ante and postnatal periods; and developing consensus on best practice for the treatment of maternal anaemia, including the dose, route and regimens of drugs including specifically iron. The role of universal supplementation strategies for iron has been evaluated in low-income postpartum women and needs wider consultation, taking into consideration all risks, including differences in rates of unintentional iron ingestion by young children in populations practicing selective versus universal iron supplementation to infants.