Because of the growing level of resistance of parasites to SP, there are concerns about how long IPTp with SP, as currently recommended by the WHO, will remain useful. Resistance to SP has been first described several years ago in East-Africa, and it is rapidly spreading in West Africa [
36]. In Benin, where IPTp has recently been implemented, the resistance rate to SP was found to be around 50% in children under five years of age living in a semi-rural area [
37].
While it is likely that SP will have soon to be replaced by a more effective antimalarial drug, it is not clear when this change will become necessary. Recently, it was shown that SP could keep its efficacy in public health applications (such as intervention on birthweight or anaemia) in spite of an impaired anti-parasite activity [
38]. In the future, the decision of switching to a more efficacious antimalarial drug than SP should be based on the association of several indicators, clinical as well as more directly related to the drug. For this purpose, an active monitoring of pregnant women in selected sentinel sites will probably be necessary.
Which indicators should be used?
Low birth weight, defined as below 2,500 g, is the main clinical indicator of the consequences of gestational malaria. It is a strong predictor of infant health [
4,
5], easily measurable and reproducible, so it can be compared between studies and countries. The only problem with LBW is that, being multifactorial, it may require very large samples to evidence differences between groups.
Maternal anaemia is usually determined at the end of pregnancy or at delivery. WHO definitions for mild and severe anaemia are a haemoglobin level below 11 g/dL and 7 g/dL, respectively [
39]. Although severe anaemia has been shown to be associated with a higher child and maternal morbidity and mortality, mild anaemia does not seem to confer unquestionably higher risks for the mother or the baby [
40]. Hence, severe anaemia should be the best indicator of the two. As the prevalence of severe anaemia is very low, a higher threshold could be used, such as 8 g/dL, as advocated by several authors [
12,
15].
Placental malarial infection is the main parasitological indicator. Compared to LBW or maternal anaemia, it is not multifactorial and depends only on the presence of parasites. Some authors have suggested to check for peripheral infections during pregnancy instead of placental infection [
41]. Nevertheless, placental infection should be more appropriate to evaluate malaria prevention strategies as it is directly linked to the newborn, whom the prevention is focused on. It has been shown to be highly correlated with late peripheral infections, which are the most deleterious for the child [
22]. Besides, women can be reached more easily at delivery than during pregnancy. To detect placental infections, thick blood smear are more appropriate than PCR and biopsy. Indeed, it is more easily feasible, cheaper, and a high correlation has been found between placental infection – determined in this way – and low birth weight [
3,
15,
17,
42] and prematurity [
20]. PCR is more sensitive to detect placental infections, but by being too sensitive it is less discriminating when comparing two drugs or strategies. Biopsy provides additional information regarding chronic and past malaria infections, but these infections are less correlated with birth outcomes, and it is more difficult to perform.
Such clinical indicators are of particular relevance as they have public health implications. However, they do not reflect the efficacy of SP-IPTp only, but the combined efficacy of all the measures that benefit to pregnant women (i.e. IPTp, ITNs and iron and folate supplementation). In order to evaluate the IPTp strategy more specifically, the monitoring of indicators more directly related to the drug (such as
in vivo SP efficacy) should be done at the same time.
In vivo therapeutic efficacy of antimalarials is usually evaluated in young symptomatic children. However, resistance rates – determined in this way – poorly correlate with public health delivery outcomes, in particular because of the incomplete immune response to malaria of children compared to adults. In a cross-sectional survey conducted in Benin, CQ was shown to be still highly efficacious to prevent LBW, while the resistance to the drug was up to 90% in the children under five years of age living in the same area [
43]. Also, a recent study discussed the relevance of extrapolating data collected in children to pregnant women, as both groups showed different sensitivity to drugs such as CQ or SP [
44]. To collect more meaningful data,
in vivo SP efficacy should be determined in pregnant women rather than children. In parallel, the correlation of this protection with molecular markers of SP resistance should be investigated.
Which alternative drugs for IPTp?
In addition, there is a need to evaluate alternative antimalarial drugs for IPTp, in terms of efficacy, tolerability and acceptability, to replace SP when it is considered not efficacious enough to protect women.
Ideal properties for an alternative drug are: (1) having a long-half life, as it has been suggested that IPTp had a prophylactic rather than a treatment effect and that the duration of prophylaxis was the most important determinant of IPTp efficacy; (2) being safe during pregnancy, and well-tolerated to ensure a high compliance with the treatment in women who are often asymptomatic when infected with malaria; (3) being easy to administer (ideally a single dose); and (4) at an affordable cost.
As recently stated by the WHO, mefloquine is presently one of the most attractive options. It has a long half-life and is still highly efficacious in African countries. Most studies have shown that MQ was safe for use in pregnancy [
3,
45‐
47]. A single study has suggested that women who received MQ treatment had a significantly higher risk of stillbirth than women not treated or treated with other antimalarials [
48]. Further studies have not confirmed this finding so far [
49]. MQ has been associated with a range of side-effects, raising the question of the women's compliance with the treatment. In an ongoing clinical trial, in which Beninese women are randomized to receive SP or MQ (15 mg/kg as a single intake) for IPTp [
35], only mild adverse events have been observed – such as vomiting, nausea and dizziness, and they resolved fast (Briand, personal communication). Very few women have refused to take the second dose because of a poor tolerance of the first intake. The drug has been well-accepted in spite of its bitter taste, and there has been less than 1% of early vomiting (within one hour) after giving a fat snack before the women took the drug (as recommended by the manufacturer).
Whatever the alternative drug, the cost will be substantially higher than for SP or CQ. Mefloquine remains expensive even if its cost has recently declined. In a near future, lower cost and higher availability of the drug in African countries should make this option even more realistic.
Artemisinin combination therapies (ACT) are also being evaluated for IPTp. They have been shown to be highly efficacious and safe during pregnancy except when used in the first trimester. However, if the effect of IPTp is mainly prophylactic [
28], then short-acting drugs would be expected to provide little benefit. Moreover, ACT are still very expensive and less easily deliverable as they require multiple treatment doses that could not be given as a directly-observed therapy in the ANC clinic. Because the treatment needs to be administered several times, compliance might be low. Other potential candidates, such as SP plus amodiaquine, SP plus azithromycine (keeping in mind the risk of pneumococcus-resistance increase when using azithromycine) and chlorproguanil-dapsone are also being assessed for IPTp. Piperaquine – used in combinations with other antimalarials rather than used alone – might be one of the most promising options for IPTp.