This study showed that if primigravidae were the most vulnerable group to anaemia at the beginning of pregnancy, they rapidly increased their Hb to reach higher levels than multigravidae at delivery. Protective interventions against malaria (i e, IPTp) appeared to play a major role in this process, which was demonstrated in three studies conducted in different areas at different times in southern Benin.
Although the women come from three different studies, they nevertheless share common factors. They all originated from the same region of Benin, with the same climatic pattern, and potentially similar malaria transmission in all sites . They were all included before 28 weeks of gestation and they had the same average gestational age at delivery. Moreover, the women did not differ in the distribution of baseline characteristics such as age, body mass index at inclusion, which are prone to influence Hb concentrations . However, although all women received two doses of IPTp with a minimal one-month interval between intakes, the study designs were different: STOPPAM being an observational study and Ouidah and APEC controlled randomized trials. The timing of IPTp also differed between the three studies, the last intake being on average one month closer to delivery in Ouidah compared with the two other studies and probably related to a lower prevalence of placental infection, as stressed by Huynh et al. STOPPAM study and APEC trial were located in more rural settings than Ouidah. Finally, investigations were held at different times and one can assume that, in particular, resistance of malaria parasites to SP may have progressed from 2005 until now.
The analyses of data from each of the three studies separately showed that primigravidae who were initially at higher risk of anaemia at inclusion (ANV1), increased progressively Hb concentration and became, although not significantly, at lower risk of anaemia at delivery compared with multigravidae. Pooling all data from the three studies increased the sample size and thus the power of the analysis, demonstrating the importance of gravidity as a determinant of maternal anaemia, even after adjusting on the characteristics of each population. In addition, taking into account gestational age in the multivariate analyses minimized the effect of potential confounders such as the gestational plasma volume changes that might as well have played a role in the parity-related difference.
Primigravidae had a lower mean Hb and an increased risk for anaemia in early pregnancy compared to multigravidae, prior to the administration of IPTp. Furthermore, an overall decrease in the proportion of malaria infections after women were given IPTp has been shown [1, 15–17]. In the study, more than 20% of primigravidae were infected by malarial parasites at inclusion, whereas only less than 10% of them were malaria positive at delivery. At the same time, the proportion of malaria-infected multigravidae decreased from 10% to 6%. The decline in the proportion of malaria infections in primigravidae also coincides with the increase of Hb concentrations on ANV2 and delivery. Such an increase of Hb concentrations in primigravidae may then be explained by the reduction of the proportion of malaria-related anaemia by IPTp in this group. These results are in agreement with a large meta-analysis of Hb parity differences comparing malarious and non malarious areas . In addition, the demonstration of a better efficacy of anti-malarial drugs in primigravidae for the prevention of anaemia had been made in the 1990s, when chloroquine chemoprophylaxis was still used . Since the implementation of IPTp with SP, two observational studies led to conflicting results [20, 21]. Rogerson et al showed in Malawian women a decrease of maternal anaemia only in primigravidae , whereas Hommerich et al showed a decrease of anaemia in only Ghanaian multigravidae . Finally, a Cochrane review of the effects of drugs to prevent malaria-related illnesses in pregnant women concluded that IPTp was effective to reduce the risk for peripheral and placental malaria and maternal anaemia, especially in primi and secundigravidae .
In spite of IPTp and iron supplementation, multigravidae increased Hb levels, but not in the same proportion as primigravidae, and consequently were at higher risk of anaemia at delivery. Indeed, in the APEC study, the proportion of women presenting with iron deficiency was higher in multigravidae than in primigravidae at each blood assessment. As previously described , there is a great demand for iron during pregnancy, as shown by the drop in Hb levels in the second trimester of gestation. Consequently, the effect of iron supplementation may be masked and even be insufficient to cover the needs of the mother and the foetus. In agreement with this hypothesis, Hernandez-Martinez et al showed in Spanish, well-nourished, pregnant women (including primi and multigravidae) who received an iron supplementation (40-60 mg/day) that iron deficiency increased importantly as pregnancy progressed (from 8% at inclusion to 68% at delivery) . A similar trend was not observed in APEC. On the contrary, iron deficiency seemed to decrease in multigravidae (from 35% at ANV1 to 31% at delivery), suggesting that the effect of iron supplementation might be better than expected. In this study, the difference between primigravidae and multigravidae may be a consequence of an increase in the prevalence of iron deficiency, due to cumulative iron requirements of successive pregnancies. Additionally, closely spaced pregnancies that are frequently observed in sub-Saharan Africa may exacerbate this phenomenon, as birth intervals have been found to affect women’s Hb concentrations, with short intervals being a risk factor for anaemia .
The lack of longitudinal data for pregnant women with the same definition of iron deficiency makes comparisons of prevalence of iron deficiency difficult among studies from sub-Saharan Africa. Nevertheless, in Kenyan pregnant women of less than 24 weeks’ gestational age, results close to the findings of this study were found by Alusala et al. The authors showed that more than 40% of multigravidae were iron deficient, defined as serum ferritin concentrations below 12 μg/L vs less than 19% of primigravidae. This is in agreement with the results of APEC (36% and 24% in the two groups).