In sub-Saharan Africa, where malaria is endemic, over 50 million women become pregnant every year. These women are twice as likely as non-pregnant women to contract malarial (plasmodium falciparum) infections (Rulisa et al, 2009). Such pregnancy-associated malaria risks are highest in women in their first pregnancy (primigravidae) and second (secondigravidae) pregnancies (Ofori et al, 2009, Mbonye et al, 2006). As this constitutes an enormous health risk to a very specific population, this review will consider the efficacy of intermittent preventive treatment of malaria in pregnancy in sub-Saharan Africa. Malaria is often asymptomatic in pregnant women who live in endemic areas. As these women are likely to have acquired adult immunity, this condition will often lead to placental malaria, maternal anaemia and low birth weight (Newman et al, 2003). In addition, malaria is now considered to be a leading contributor to maternal mortality among pregnant women in sub-Saharan Africa (Mbonye et al, 2005). Malaria also accounts for approximately 2-5% of maternal anaemia, 8-14% of cases of low birth weight and 3-8% of infant mortality (Rulisa et al, 2009) In order to prevent malaria, The World Health Organization (WHO) recommends a weekly or bimonthly administration of anti-malarial chemoprophylaxis, consisting of either chloroquine (CQ), sulfadoxine-pyrimethamine (SP) or dopson-pyrimethamine (Briand et al, 2008). These prophylaxes (chloroquine the most commonly used) were effective in reducing low birth weight, placental and maternal malaria. Due to non-compliance by mothers and the increasing resistance of malaria parasites, WHO has also published guidelines which changed its malaria treatment recommendations to include intermittent preventive treatment (IPT) (Briand et al, 2008). Intermittent preventive treatment consists of a single dose of anti-malarial drug administered at least twice in pregnancy, typically during the second and third trimesters. This treatment also includes routine checkups at antenatal clinics (ANC) irrespective of whether the mother is parasitemic or not. Sulfadoxine-pyrimethamine is the recommended drug for this treatment (Akinleye et al, 2009, Newman et al, 2003). While IPT can be best obtained at these antenatal clinics, some malarial mothers will not attend these clinics, or fail to return after their first appointment. This patient noncompliance tends to make the administration and monitoring of IPT regimens difficult for doctors and caregivers (Brabin et al, 2009, Hommerich et al, 2007) The objective of this review is to examine the efficacy of current recommended intermittent preventative treatment using sulfadoxine-pyrimethamine (IPT/SP) in reducing complications and improving pregnancy outcomes in malarial-endemic sub-Saharan African countries. IPT/SP strategy is recent, introduced in 2000 and implemented in some hard-hit nations only as recently as 2003 (Akinleye et al, 2009). This paper analyzes studies published during the period 2003-2009. This review is divided into three parts. The first part will analyze perceptions, knowledge, and practices regarding the effects of malaria in pregnancy. The second part will compare chloroquine (CQ) to sulfadoxine-pyrimethamine (SP) to establish the efficacy of IPT. The third part will explore uptake and timing of IPT/SP by pregnant women in sub-Saharan Africa.
Malaria in Pregnancy
Mbonye et al (2006) conducted an exploratory study in mukono-Uganda, which focused on perceptions of malaria in pregnancy, recognition of symptoms and perception of the seriousness of malaria and the effect of culture on treatment and prevention of malaria. The exploratory study consisted of 10 focus group discussions (FGD) and 40 informant interviews. Of the study participants, 36 were women between the ages of 20 and 49, and 15 were adolescent girls between the ages of 10 and 19. The remaining 39 participants were men between the ages of 20 and 50. The study found that this group perceived malaria and anaemia to be the most common illness associated with pregnancy, and though they associated anaemia with poor nutrition, low birth weight was also seen as a consequence of anaemia. The study respondents also identified pregnant women and infants to be especially susceptible to malaria. In addition, these respondents perceived malaria as present in a mild and a severe form. The mild form was perceived as consisting of a low grade temperature, joint weakness, nausea, loss of appetite, a sour taste in the mouth, general lethargy and palpitations. The severe form of malaria was identified by the respondents as consisting of high fever, convulsions, confusion, rigors, vomiting and a yellowness of the eyes. Some of the respondents also associated malaria with miscarriage and stillbirth yet insisted that pregnant women could not have malaria without showing any of the aforementioned symptoms. The study respondents did not have knowledge of asymptomatic parasitemia in pregnancy or the concept of IPT. This study also identified that the signs of early pregnancy and the symptoms of malaria were unclear, especially to the pregnant women in the study. Fever in pregnancy was also perceived as normal, while malaria in primigravidae and adolescents were seen as low risk. The risk of disease was exacerbated by how it was seen as culturally acceptable to hide pregnancy until the mother was sure. This can be explained by how pregnancy outside of marriage was frowned upon among all respondents, causing women who suffered illness in pregnancy to not openly seek help. The study also found strong use of herbs and folk medicine to treat illness in pregnancy, which respondents attributed to an avoidance of cost and fear of side effects associated with modern medicine. Though this study provided some strong insight with regard to designing public health interventions, it is exploratory and not representative of the general population. Ofori et al (2009) conducted a longitudinal study in Ghana which assessed the incidence, prevalence and clinical markers of pregnancy associated malaria. This study consisted of broad-based questionnaires bolstered by blood monitoring. 294 women were recruited in early pregnancy and were surveyed until delivery. 39.5% were primigravidae, 31.6% were secondigravidae and 6% had traits associated with sickle cell anaemia. Participants were surveyed weekly to assess their malaria morbidity, and these results were corroborated by study of placental tissues and microscopic analysis of blood samples. In addition infant birth weight was documented to add to the body of evidence. This study found that haemoglobin levels were lowest in primigravidae and highest in the para-gravidae. The prevalence of placental parasitemia and low birth weight was associated with parity and was higher in the primigravidae stage of pregnancy than it was in the paragravidic. Over the course of the study participants who developed symptoms of malaria were treated with CQ. The study found that the placental malaria prevalence, at 35%, was lower in those who had developed clinical malaria than those who had not. This reduction in placental parasitemia supports the goals of IPT with regard to reducing accumulation of malaria parasites in the placenta. Though this can be seen as promising news, a 35% prevalence level is still extremely high, especially as it follows the anti-malarial treatment. These results can be seen as proving the ineffectiveness of CQ in treating malaria in pregnancy. These researchers suggested that women with placental malaria had increased risk of having a baby with a low birth weight. In addition, decreased haemoglobin levels were seen as an indication of placental malarial infection. These authors failed to control for the effect of nutrition and socio-economic status on anaemia in pregnancy. Analysing samples microscopically instead of histologically did not provide enough evidence to support claims that decreased haemoglobin indicated placental malaria. These results do refute the conclusions of Mbonye et al (2009) that there is a low risk of malaria in Primigravidae pregnancy. Newman et al (2003) conducted a cross sectional study in Ethiopia to determine species-specific malarial rates during pregnancy and the associations on pregnancy outcomes between areas of stable transmission (ST) and unstable transmission (UT). These researchers recruited 962 women from antenatal clinics and delivery units (DU), delivering a questionnaire which focused on socio-demographic data. These results were combined with laboratory blood analysis of samples provided by the participants. This study found that malaria in pregnancy was associated with low birth weight, anaemia and premature stillbirth. The prevalence of malaria and anaemia in the area of stable transmission was lower than what was found in the area of unstable transmission. In addition, the women in the study who were parasitemic in the unstable transmission area were found to be more likely to present fever and to suffer from moderate to severe anaemia. This study also identified an association between increasing gravidity and a decreasing rate of peripheral and placental parasitemia. Apart from P. Falciparum, these researchers found 19% of the women in the unstable transmission area infected with P. Vivax as well. This study was conducted in an urban area during a non-epidemic season. In addition, both ANC and DU were easily accessible. For these reasons the results of this study cannot be considered a true reflection of the burden suffered by pregnant women in rural areas during malarial epidemic seasons. In addition, the actual association between P. vivax and placental parasitemia was not assessed by these researchers. However, these findings align with those established by Ofori et al (2009), whose study also identified placental parasitemia to be strongly associated with parity and low birth weight. Although these studies have successfully linked placental malaria with low birth weight, other studies have proven otherwise. For example, an observational study by Rulisa et al (2009) compared pregnancy outcomes in different areas of Rwanda with endemic malarial outbreaks. This study consisted of analysis of birth data from 12, 526 infants from eleven delivery centers. The study showed that malaria incidence had no significant effect on incidents of low birth weight among infants, but the infection had an immense seasonal effect on birth weight (BW). Evidence for this phenomenon was derived from recordings of birth BW in different seasons: Recorded BW was highest in the period beginning with June and ending in September and lowest in the period beginning in October and ending in February These researchers found that the increases in healthy BW coincided with the harvesting season, a time when food was in abundance. In addition, these researchers concluded that the improvement in socioeconomic and nutritional status could play a role in BW, as evidenced from the increase in BW during these periods. These findings contrast the findings of Ofori et al (2009) and Newman et al (2003) which found malaria in pregnancy to be associated with low birth weight, but did not take into account seasonal and socioeconomic factors. Data collected by Rulisa et al. was also based solely on data collected by primary care facilities, and these findings alone cannot prove the accuracy of the data. Though these findings may show a low transmission rate of malaria, specifically in Rwanda, they should not be used as a strong basis for recommendations that pregnant women in malaria-prone areas are protected against malaria.
Efficacy of IPT/SP and CQ chemoprophylaxis
In a randomised control trial, Kayentao et al (2005) compared three treatment regimens for efficacy in malarial prevention in primigravidae and secondigravidae women in Mali. This study sample comprised of 1,163 women in their first or second trimesters (between 16-26 weeks of gestation). These participants were divided into three groups: The first group received a weekly CQ chemoprophylaxis at the first visit to an antenatal clinic; the second group was given two doses of IPT with chloroquine at enrolment and again in the third trimester; the last group was given two doses of IPT with SP at enrolment and in the third trimester. All participants in this study were advised to not self-administer any anti-malarial drugs apart from those prescribed under the study regimen. At delivery, these researchers collected blood samples, neonatal weight and gestational age data. This study found parasitemia among study subjects to be lower in those who were given IPT/SP than those who had a regimen of CQ and IPT/CQ. The proportion of women who developed third trimester anaemia was higher in the group administered weekly CQ and IPT/CQ than those administered IPT/SP. Those in the CQ and CQ/IPT group were also found to be more likely to develop placental malaria, stillbirths, infants with low birth weight, and abortions. Placental malaria was also strongly associated with LBW and maternal anaemia. This study emphasises the ineffectiveness of CQ and corroborates the findings of Ofori et al (2009). Of the original 1,163 studied, only 1,110 participants were followed up to delivery. The author does not provide a suitable explanation of how remaining 95 participants were lost and whether or not these were replaced. In a cross-sectional study, Briand et al. (2008) assessed the efficacy of CQ chemoprophylaxis and IPT/ SP against accounts of low birth weight in Benin. Of the 1,699 women in the study, 1,090 comprised the CQ group and 609 comprised the IPT/SP group. This study found out that the mean birth weight in CQ group was 2876g, as compared to 3060g in the IPT/SP group. Accordingly, these researchers concluded that the prevalence of placental parasitemia was significantly higher in the CQ group, and extrapolated that IPT/SP is more effective than CQ in preventing low birth weight and placental malaria. These findings are consistent with Kayentao et al (2005). However, the findings of Briand et al. (2008) could have been more persuasive and valid if a control group was introduced for clarity. Falade et al (2007) evaluated the epidemiology of congenital effects of malaria in Nigeria, as compared to the effect of IPT/SP and Pyrimethamine (PYR) as prophylaxis in preventing maternal and placental malaria. Out of the 983 mothers which were examined in this study, 60.8% received IPT/SP, 21.8% received PYR and 17.4% did not receive chemoprophylaxis (NC). Most of this study's participants also used a vector control measure like window nets or mosquito spray. 84.4% of those in the IPT/SP group reported receiving at least two doses per guidelines whereas 94.4% of those in the PYR group claimed to have taken adequate doses. This study found considerably less maternal parasitemia in the IPT/SP group than the PYR and NC group. Maternal anaemia and low birth weight among infants born during the study period was lower among mothers in the IPT/SP than those in PYR and the NC group. Those in IPT/SP were less likely to have premature births. These findings indicate that IPT/SP is efficient in reducing placental and maternal malaria and is consistent with findings from Mali, Benin, Nigeria and Ghana. The use of other vector controls did not alter the outcome of the study as the pattern of vector control was similar in both groups. However, 60.8% of women in the IPT group is too significant a difference from the other variables to bear the same characteristics as the other groups. Moreover, these researchers failed to explore how vector control might have contributed in the reduction of maternal and placental anaemia. In addition, it seems that the PYR group was not monitored in their uptake of the anti-malarial medication as it appears participants self reported their pyrimethamine intake. That such crucial information was left to the patients to report raises serious questions of compliance and validity of results
Timing and IPT uptake.
Brabin et al (2009) conducted a qualitative study to assess the awareness of rural Gambian women on the timing and the safety of their two-dose IPT schedule. This study consisted of 16 focus group discussions and 41 interviews with men, women, adolescents and traditional birth attendants. The results of this study showed confusion regarding the relevance of IPT schedules among the rural Gambian population. This confusion manifested as such: Women conceptualised their pregnancy in stages and not in weeks or trimesters hence had difficulty in understanding the proper timing of IPT. In addition, subjects showed considerable confusion regarding the different drugs necessary for pregnancy administration, such as haematinics. Due to this confusion, pregnant women in this study would fail to distinguish the efficacy of IPT from other treatment methods and thus were entirely dependent on the staffs of antenatal clinics for safe IPT administration. Qualitative results are not generalised but results like these can be effective and influential with regard to developing programmes to assist mothers who lack sufficient information regarding modern pregnancy staging in calculating the appropriate time of administering IPT. This study did not account for determinants of IPT uptake based on perceptions of safety. Anders et al (2008) conducted a study which focused on the determinants and uptake of IPT by pregnant women in Tanzania. In a study of 133 pregnant women, 75% attended an ANC in their first trimester, 93% attended at least once before the third trimester and 7% whose first visit was in the third trimester attended due to a lack of problems with their pregnancy. Most of the study respondents were given IPT after six months of gestation but 13% reported receiving treatment at an earlier period. Nearly half of the subjects attended ANC during the fourth month of pregnancy but did not receive IPT on their first visit. This phenomenon was due to policies which restricted doctors to giving the doses only in the second and third trimester. Since some mothers did not return after the first visit, this practice resulted in a missed opportunity to maximize the benefits of IPT. These researchers also found that ANC alone could not improve uptake of IPT, singling out factors such as insufficient stock and lack of information as those which had the most negative impact on uptake. This study was limited by two factors: the sample size was too small to generalise the result beyond the population the sample was drawn from, and the sample were not randomly selected. In a cross sectional study which assessed knowledge, practices and attitudes with regard to IPT services and malaria control in north-eastern Tanzania, Mubyazi et al (2005) interviewed health officers, dispensary-based staff and pregnant women in focus group discussions. A total of eleven managers took part in the study as key informants. Study participants were interviewed regarding their views on how to improve ANC services with regard to IPT for pregnant women. This study found that there was an overall lack of comprehensive knowledge regarding malaria prevention and management among mothers who attended antenatal clinics. The women interviewed showed concern about the use of SP in pregnancy, expressing apprehension regarding the side effects on unborn babies. Corroborating this finding, these researchers observed that pregnant women would often discard SP tablets once they left the health centre. Most of the study participants agreed that some implementation of direct observation therapy could improve uptake. Staff also admitted that since IPT is a relatively new concept, not all healthcare staff were fully knowledgeable on the topic. The fear expressed by participants for the side effect of anti-malaria is consistent with results found in Uganda by Mbonye et al (2006). Another study by Mbonye et al (2008) assessed pregnancy outcomes as well as the delivery of IPT through community resource volunteers, comparing these results against similar provisions provided in the healthcare unit. This study was conducted in 25 parishes in Uganda; with 21 of the parishes using the community approach to administer IPT and 4 used the health units approach. 51 people were identified and trained to be able to deliver IPT in the community, and 2785 women were recruited to participate in the study. Study data was collected by the community resource volunteers at recruitment before the women received the doses of IPT. Study results showed that women who access IPT through the community delivery system had higher rates of use and adherence to an IPT regimen due to the greater ease of access through the community program. In both approaches, though, there were marked reductions in rates of anaemia, episodes of malaria and low birth weight. These results support similar studies which found IPT to be effective. This study is not without its limitations. Relying on data collected by the community resource volunteers could pose a great limitation to this study as the community based volunteers are not professionals, though they received some training. In addition, no clinical or laboratory tests were performed to measure incidence of placental malaria during this study. This lack of information may have lead to an overestimation of malaria incidence and perhaps the indiscriminate administration of malaria medication in pregnancy. Torimo (2007) analyzed the incidence of malaria and anaemia during pregnancy among 395 women who attended a routine antenatal clinic in Kibaha district in Tanzania. In a descriptive cross sectional survey, this researcher explored factors which might influence IPT/SP uptake. Blood from the subjects was tested and parasitemic treatment was given, but if the subjects' pregnancies were between 20-28 or 30-38 weeks then IPT/SP was administered. The mothers were then interviewed on their knowledge of anaemia and malaria control in pregnancy. The results of this study showed that of the 395 pregnant women 27.3% were parasitemic, 34.2% had severe anaemia with haemoglobin levels(HB) of less than 8g/dl and 56% had moderate anaemia(HB between 8.0-10.9g/dl). Mothers who did not receive IPS/SP due to unavailability constituted 40% of those surveyed. These results are consistent with other findings in Tanzania which concluded that a lack of stock had a profound effect on uptake. 90% of the surveyed mothers knew about IPT/SP treatment but 70% did not know about the necessity of properly timing IPT/SP treatment according to gestational period. In addition, 40% of those surveyed preferred to swallow the tablets at home, explaining that they did not want to share cups with other patients. This finding raises serious doubts about patient compliance. The recruitment criteria were also inappropriate since the subjects were only interviewed upon exiting the clinic. The researcher may not have had enough time to seek informed consent, but this sample nonetheless cannot be a true representation of the population. Furthermore, the researcher failed to describe how he followed up with these patients.
Nearly all the studies covered in this review have shown IPT/SP to be efficacious in malarial treatment among pregnant women in sub-Saharan Africa. In addition, these studies have shown that mothers ought to have access to greater education regarding the importance of gestational timing in IPT regimens as to improve uptake. Further, these studies have shown the need for greater research to be done regarding the socio-economic causes of low birth weight.