Severe anaemia is a frequent cause of admission to hospitals in tropical Africa and about 10% of such children die. In endemic countries, anaemia has multiple causes such as nutritional deficiencies, infections and haemoglobinopathies. However, Plasmodium falciparum infection is believed to be the major contributory factor to the aetiology of severe anaemia. Severe anaemia is usually treated by blood transfusion although transfusion carries the attendant risk of transmission of HIV and other blood-borne infections. Thus, there is a need to explore novel strategies to reduce the incidence of severe anaemia in high-risk groups such as children with suboptimal haemoglobin levels because these children are at increased risk of developing severe anaemia if they develop a malaria infection before their haemoglobin level has normalized. Therefore, it is proposed to study whether monthly chemoprophylaxis with sulphadoxine/pyrimethamine (S/P) given during malaria transmission season can protect Gambian children from developing severe anaemia. After receiving treatment from the hospital, 1200 children admitted to the hospital with a haematocrit of less than 21% were randomised to receive either monthly S/P or placebo during the rest of the malaria transmission season. Morbidity was monitored throughout the rainy season. Study subjects were seen at the end of the dry season to document morbidity and mortality.

Objectives

1. To assess the effectiveness of monthly prophylaxis with S/P on haematological recovery in children who have been treated for severe anaemia.
2. To measure compliance with the intervention
3. To investigate whether or not infection with chloroquine and S/P resistant strains contributes to the pathogenesis of severe anaemia

Study design and methods

Study site:

The study was based at the 75-bed paediatric unit of The Royal Victoria Hospital (RVH), which is located in the capital, Banjul. This is the main referral hospital for the country. The paediatric unit admits about 7000 children a year, of whom about 800 are anaemic (PCV < 33%).

Enrolment of children:

Study subjects were recruited fom paediatric unit of RVH, MRC Hospital, and Brikama, Essau and Sibanor Health Centres. Study subjects were enrolled by non-coercive methods if they fulfil the inclusion criteria and had none of the exclusion criteria.

Screening process and enrolment:

Patients were recruited on presentation to the ward. All children in the right age group who presented to the ward had their haematocrit and blood film done by a project field worker. Children with a PCV of less than 21% were offered assessment by a project doctor. If assessment was permitted, and the patient fulfilled the entry criteria, then written consent was sought from the parent/guardian. The study was explained in the parent's/ guardian's preferred language, and written informed consent was obtained on approved forms.

On entry into the trial a detailed clinical history, drug history, family history and information on bednet usage and impregnation was obtained. The study physician examined the child, documented nutritional status and recorded the findings in a case report form. In addition, a detailed address was obtained. Each study child was provided with a photo ID card to facilitate identification at each contact.

A 5-ml baseline blood sample was obtained for determination of the following:

• Hb, red cell indices, white cell and platelet count (Coulter counter)
• Reticulocyte count
• Markers of iron status (serum iron and ferritin) (first 100 children)
• Sickling test and, if positive, haemoglobin electrophoresis
• Stool examination for hookworm and other parasites (first 100 children)

Other investigations such as chest x-ray and urine examinations were carried out as clinically indicated.

Inpatient management:

During admission, children with malaria were treated with IM quinine or chloroquine plus S/P as clinically indicated. Children with a PCV < 15% and those with signs of respiratory distress and/or a 'gallop' rhythm were transfused. In addition, all patients received iron for 28 days, treatment starting at the time of their discharge from hospital. Patients were treated for other conditions as clinically indicated.

Treatment regimens

Treatment was given as follows:

S/P (tablets containing 500 mg sulfadoxine/ 25 mg pyrimethamine) at an approximate dose of 1.25 mg pyrimethamine/25 mg sulphadoxine per kg stat dose.

Oral chloroquine: 10mg daily for 3 days.

S/c chloroquine: 5mg/kg 6 hourly x 7 doses.

IM Quinine: 20 mg/kg loading dose, then 10 mg/kg 12 hourly for 5 days.

Iron was given as syrup at a target dose of 2 mg/kg elemental iron.

Assessment of resistance to chloroquine and antifolates :

To investigate whether or not infection with chloroquine or S/P resistant parasites contributes to the pathogenesis of severe anaemia in Gambian children, filter paper blood samples were collected from the first 100 children entered into the study who have P.falciparum parasitaemia and an equal number of aged matched children who attended the OPD clinic with uncomplicated symptomatic malaria. Nested PCR on extracted DNA was used to tested for mutation in genes associated with resistance; Pfcrt and Pfmdr for chloroquine and DHRF / DHPS genes for S/P using standard techniques.

Study subjects were individually randomised into either the S/P or the placebo group in a 1:1 fashion at the time of discharge from hospital, although chemoprophylaxis was not be started for another week as the children are likely to have received antimalarials whilst in hospital. Randomisation was in blocks. Block size were selected to enhance blinding and maintain balance in the treatment allocation ratio as recruitment progresses

Home visit:

As soon as possible after discharge from hospital, patients were visited at home by a project field worker to collect information on social economic and environmental risk factors that may predisposes an individual child to develop severe anaemia. A housing survey was conducted during this visit to determine possible risk factors for malaria which might act as confounders in determination of the primary trial end-point. Field workers interviewed the parents/guardians of study subject to record characteristics about the house that might influence mosquito biting behaviour. These items included building material, an evaluation of the presence of eaves, impregnation and condition of bed nets. During the visit, parents were encouraged to protect the study child with a bed net.

First OPD visit after discharge from hospital One week after discharge, study subjects enrolled in the study were asked to return to the OPD clinic for routine assessment. At this visit a fingerprick blood sample was collected for PCV measurement and thick blood film for malaria parasite. The first dose of trial medication was given during this visit.

Randomisation:

S/P and matching placebo (identical shape and size) were stored in 8 boxes labelled 1 to 8 (4 for S/P tablets and 4 for placebo). The randomisation code were kept in sealed opaque envelopes by the local safety monitor who alone will knew the code, which associated each treatment group with either S/P or placebo.

Continuation of chemoprophylaxis in the health centre:

Study subjects received their monthly chemoprophylaxis at their local health centre under direct supervision by staff of the health centre identified by the health centre and trained to take this important role in the study. Health centre staff were provided with a list containing the study number, photographs and randomisation group of each study subject. Study subjects were observed for 30 minutes after dosing and if vomiting occurs within 30 minutes, the dose was repeated. S/P (tablets containing 500 mg sulfadoxine/ 25 mg pyrimethamine) will be given at an approximate dose of 1.25 mg pyrimethamine/25 mg sulphadoxine per kg. Enrolment of patients started in July and surveillance continued until end of malaria transmission season.

Morbidity surveillance during the rainy season:

Study subjects were followed throughout the rainy season. Mothers/guardians were encouraged to take their child to the RVH OPD clinic or the health centre identified as being closest to their home at any time after discharge if the child became unwell. Project staff were based at each of these health centres, initially to identify children in the trial and to ensure that they were seen, properly investigated and treated promptly by health centre staff. At each visit axillary temperature was recorded using a digital thermometer. A dipstick for diagnosis of malaria was used to guide treatment if fever (axillary temperature of > 37.5°C) or history of fever within the previous 48 hours is present. In addition, a thick blood smear for malaria parasites was collected for subsequent confirmation of the diagnosis. Filter paper sample to test for S/P resistance were collected from all malaria cases detected during morbidity surveillance. Study subjects with documented fever (axillary temperature of > 37.5°C) or history of recent fever and malaria parasitaemia were treated with S/P and chloroquine. The treatment of study subjects seen at the health centres was the carried out by health centre staff in accordance with national guidelines. Study patients who needed admission were referred to RVH paediatric unit or the MRC ward.

End of malaria transmission season cross-sectional survey:

Children enrolled in the study were seen at the OPD clinic at the end of malaria transmission season for examination by a study physician and a finger-prick blood sample was obtained for preparation of thick blood smear and determination of haematocrit. A standardized questionnaire was administered to the parents/guardians of the study subject, to collect information regarding illness that had occurred since the last visit, symptoms experienced, use of healthcare facilities and use of medicines. Information on the use of bednets was collected again at this visit.

Dry season follow-up:

At the end of the dry season study subjects visited once. During this visit, a questionnaire was administered to document morbidity and mortality.

Sample size calculations:

Assuming an attack rate of 0.1 episodes of clinical malaria (as defined for the primary end-point) per month in children in the control group (a figure based on previous surveys from The Gambia) and a drop out rate during follow up of 15%, 131 episodes was expected in children in the control group followed until the end of one malaria transmission season. Thus, with a total of 1200 children recruited to the trial, the study will have an 80% power to detect at 5% level of significance a 32% reduction in incidence of clinical attacks of malaria in children who receive chemoprophylaxis.

Previous studies in The Gambia indicate that about 20% of children under 5 years of age have a PCV of < 20% at the end of the malaria transmission season. With 1200 children enrolled, this trial will have a 80% power to detect a 34% reduction in the prevalence of anaemia of this degree of severity.

Inclusion Criteria:

1. Age: 3 months to 9 years.
2. Haemoglobin concentration 7g/dl on admission to hospital.
3. Residence within 30 km of the study centre and availability for the duration of the study period.
4. Informed consent to participate in the study given by the parent or guardian.

Exclusion Criteria:

1. Simultaneous participation in any other trial.
2. Allergy to sulpha drugs.
3. Residence > 30 km from the recruitment site.
4. Lack of consent.