Malaria in pregnancy is potentially fatal to both the mother and the foetus particularly in the primigravidae. Implementation of appropriate control and preventive measures is challenged by the fact that malaria infection in pregnancy is often asymptomatic and parasitized red blood cells sequestrated in the placental microcirculation may not be detectable in the peripheral blood. In addition, the widespread prevalence of parasites resistant to chloroquine and sulphadoxine-pyrimethamine (SP) and, the safety concerns about newer antimalarials, poverty and inadequate supply have made antimalarial treatment options available to pregnant women very limited. These have necessitated an urgent search for alternative safe and efficacious treatment options for pregnant women. The objective of this study is to assess the efficacy, safety and tolerability of four antimalarial treatment options in rural Ghana within a programme setting.

Primary objective:

To determine the effect of AQ, SP and the AQ+SP combination compared with CQ treatment on the prevalence of peripheral parasitaemia on days 14 and 28 post treatment.

Secondary objectives:

1. To compare the incidence of adverse events in the treatment groups.
2. To compare the effect of study drugs on maternal haemoglobin on 14 and 28 days post treatment, and at delivery.
3. To compare the effect of study drugs on peripheral and placental parasite densities at delivery.
4. To compare the effect of study drugs on birth weight at delivery.
5. To assess the effect modifications of gestational age, parity, gravidity, prior antimalarial use, presence or absence of symptoms at enrolment, baseline parasite density and baseline Hb on the parasitological and haematological responses to the test drugs.
6. To assess the accuracy of the OptiMAL antigen test for detecting peripheral parasitaemia compared to microscopy.
7. To compare the overall incidence of adverse pregnancy outcomes (abortion, stillbirth, congenital abnormality, prematurity and intrauterine deaths) in the study group to local rates obtained from St Theresa's Hospital's records.

Study location and population:

The study was carried out at the St. Theresa's Hospital in the Nkoranza district of the Brong Ahafo Region of Ghana. The St. Theresa's hospital is a general district hospital. It has a bed capacity of 80 and provides all basic medical services including adult medicine, paediatrics, surgery and obstetrics and gynaecology. The study enrolled pregnant women of all parities attending the St. Theresa's Hospital's antenatal clinic with a gestational age of 16 weeks and above between March 2003 and September 2004.

Methods:

Antennal screening and enrolment:

All pregnant women who attended antenatal clinics were screened for malaria antigens with OptiMAL dipsticks. Those with a positive antigen test were considered eligible, and after informed consent had been obtained from them 5mls of venous blood was drawn from an antecubital vein for baseline measurements of haemoglobin, white blood cell counts (total and differential), bilirubin, alanine aminotransferase, aspartate aminotransferase and gamma-glutamyl transferase and for making filter paper blood spots. Women were then assessed clinically and obstetrically with the view to enrolling them into the study. Pregnancy viability and gestational age were confirmed with ultrasound scanning by the study clinician or the principal investigator. Pregnant women with positive malaria antigen tests confirmed microscopically were randomised into four treatment arms if they satisfied all inclusion criteria.

Follow-up schedule:

Field workers visited the study women in their homes following the initial supervised drug administration at the antenatal clinic on post treatment days 3, 7, 14 and 28 and performed the following routines.

• Day 3 and Day 7: Obtained venous blood for filter paper blood spots, white cell and malaria parasite counts, and recorded any side effects.
• Day 14 and Day 28: Obtained venous blood for white cell and malaria parasite count, measurement of alanine and aspartate transaminases, bilirubin and for filter paper blood spots and recorded any side effects.

Subsequently, pregnant women were seen at the antenatal clinic monthly and, for those with 32 weeks and above of gestation fortnightly. At these visits, they were actively screened for peripheral parasitaemia using OptiMAL dipstick test. At any time before delivery if the test was negative, the woman remained on daily haematinics. If women who were already enrolled had positive antigen test confirmed by microscopy, they received another course of the treatment they were initially assigned to. Women were enrolled in the study only for the first episode of malaria detected during the antenatal visit. At delivery, midwives recorded birth weights and any stillbirths, perinatal deaths or congenital abnormalities. They also made slides from peripheral, placental and cord blood and sampled maternal blood for haemoglobin measurements. Any record of a congenital deformity was verified and confirmed by a clinician. The women and their babies were visited at home at six weeks post delivery to record any neonatal adverse events such as deaths or severe morbidity.

Outcome measures:

Primary

1. Prevalence of parasitaemia on days 14 and 28 post treatment.

Secondary

1. Incidence of adverse drug events within seven days following treatment.
2. Proportions of pregnant women withdrawn from the study due to the occurrence of adverse drug events (clinical and laboratory) by day 7 following initiation of treatment.
3. Change in maternal haemoglobin concentrations at days 14 and 28 following treatment.
4. Prevalence of peripheral parasitaemia at delivery.
5. Prevalence of placental parasitaemia at delivery.
6. Proportions of abnormal biochemistry and white blood cell values on days 14 and 28 post treatment.
7. Sensitivity, specificity, positive and negative predictive values, likelihood ratios, and the area under receiver operating characteristic (ROC) curve for the OptiMAL antigen test.
8. Incidences of adverse pregnancy outcomes in the study group.
9. Prevalence of postpartum parasitaemia.
10. Prevalence of postpartum anaemia.

Sample Size:

This was based on the assumption of a 28-day parasitological clearance of 90% for AQ, SP and the AQ+SP combination, and 78% for chloroquine (α = 5% power = 90%). Allowing for a 15% loss to follow-up, 225 pregnant women were recruited into each of the 4 treatment arms of the study giving a total study size of 900 pregnant women.

Data and safety monitoring board:

A data and safety monitoring board (DSMB) was constituted for the project. The board was responsible for:

• Regular monitoring of the data and safety issues concerned with the study.
• Reviewing the PI's reports on serious adverse events and making recommendations on further progress of the study.
• Reviewing the statistical analysis plan prior to breaking the study drug codes.

Inclusion Criteria:

• Gestational age of at least 16 weeks.
• P. falciparum parasitaemia of any density with or without symptoms.
• Informed consent.
• No known adverse reaction to any of the study drugs.
• Residence in the study area.

Exclusion Criteria:

• Past obstetric and medical history that might adversely affect the interpretation of outcomes such as repeated stillbirths and eclampsia.
• History of severe adverse drug reactions to co-trimoxazole in the past.
• Haemoglobin concentration below 5.0 g/dl.
• Severe malaria.