1 Introduction
Malaria is highly prevalent in Nigeria and accounts for approximately 40% of global malaria mortality. It remains an important cause of childhood morbidity and mortality globally and in sub–Saharan Africa. Approximately 80% of all malaria cases globally [
1] and 90% of malaria-related mortality occur in sub–Saharan Africa. Malaria mortality is caused by the severe, complicated forms of infection with
Plasmodium species [
1]. In sub–Saharan Africa, most severe malaria cases are caused by
Plasmodium falciparum, which is the major cause of severe malaria globally [
1].
Severe malaria morbidity and mortality are grossly underestimated because of the paucity of information from areas where the disease is most severe. The sources of information are mostly case reports and a few hospital-based studies [
2] at the national and sub-national levels, which are often incomplete. Modelling and surveys based on limited data points are not meaningful [
2]. The challenge of accurately describing malaria epidemiology is important, considering that most mortality from malaria occurs in the poorest communities with the least access to health services [
3].
Historically, in 1987, 90% of malaria mortality in The Gambia occurred in the community according to Greenwood
et al. [
4]. However, this figure has remarkably decreased over time, in which 47% was reported in Lusaka in 2011 by Mudenda
et al. [
5]. The accuracy of morbidity and mortality data are critical to the planning of interventions aimed at malaria control. The factors influencing the manifestations of infection and progression to severe disease include the infecting parasite species, the levels of innate and acquired immunity of the host and the timing and efficacy of any treatment administered [
2].
Malaria endemicity measures the intensity of transmission and determines the disease burden and clinical pattern of the disease, making the prediction of mortality from malaria endemicity possible. Our study aimed to describe malaria endemicity in the local government area (LGA) of Wamakko, Sokoto State, Nigeria using the prevalence rate of
Plasmodium falciparum parasitaemia among children aged 2–10 years (PfPR2-10) and to report the prevalence and patterns of severe malaria in the same study [
6].
2 Materials and methods
This paper presents an analysis of severe malaria, excluding malaria cases with severe anaemia. This cross-sectional community-based study aimed to determine malaria endemicity using PfPR2-10 in the LGA of Wamakko, Sokoto State, Nigeria and proposed a different method of mapping the transmission intensity and the proportion of parasite species causing clinical malaria. The methods used in the original study and the results of prevalence and mapping were published previously [
6].
2.1 Ethical statement
Ethical approval was obtained from the independent ethics committee of Usmanu Danfoidyo University Teaching Hospital (UDUTH), Sokoto, Nigeria (approval number UDUTH/HREC/2014/No. 246) and the independent ethics committee of Sokoto State Ministry of Health (approval number SMH/1580/VIV). Permission to visit was also obtained from the district heads of the involved communities. Written informed consent was obtained from the parents or legally accepted representatives of each child included in the study.
2.2 Study setting
The study was conducted in the LGA of Wamakko, Sokoto State located in the northwestern geopolitical zone of Nigeria.
The average rainfall per annum is approximately 1000–2000 mm with prominent rainy and dry seasons. Malaria transmission occurs all year round, and the peak malaria transmission in this area is observed in August. The area is mesoendemic for malaria based on an observed PfPR2-10 of 34% in our study [
6]. PfPR2-10 is a metric used to describe the intensity of malaria transmission in a geographic location.
2.3 Study design
The study was a two-point, cross-sectional, prospective descriptive study conducted during the rainy and dry seasons in August and December 2016, respectively.
2.4 Sampling technique
Multistage cluster sampling was employed in proportion to the sample size. Approximately 892 households were required to meet the target of at least 500 participants per season based on the assumptions of a 70% response rate and that 80% of households include at least one child less than 5 years of age in accordance with the 2010 Nigeria Malaria Indicator Survey (MIS). All children in the secondary clusters who fulfilled the inclusion criteria and whose parents consented to participate were included in the study.
2.5 Sample population
The participants were visited at home. They were provided with information regarding the study, and all eligible children within the household were invited to participate. The parents and primary caregivers who accepted to have the children in the household included in the study signed or thumb-printed the informed consent form. All children in the selected settlements who met the age criteria of 2–10 years were recruited for the study.
2.6 Procedures
The study procedures were performed at a central location in each of the study villages. The World Health Organization (WHO) clinical criteria was used in diagnosing severe clinical malaria [
7]. The criteria included cerebral malaria, multiple convulsions, dehydration, prostration, persistent vomiting, jaundice, and pulmonary edema. Blood samples were collected for malaria microscopy. Any child with two or more features or cerebral malaria was diagnosed with severe malaria. Hyperparasitaemia was diagnosed based on a parasite count of 250 000/µL or 5% of parasites parasitised. Concomitantly, malaria was rapidly diagnosed using CareStart® Malaria HRP2 rapid detection tests (Access Bio, Inc., model G0141), which can detect
P. falciparum. Haemoglobin was not measured in this study based on the design and aim of the original study; thus, unable laboratory-related criteria for severe malaria were not assessed. Children with severe malaria were treated with an initial intramuscular dose of artesunate administered according to age [
7] and immediately referred to the nearby tertiary hospital UDUTH in Sokoto.
2.7 Statistical analysis
Data were analysed using Stata version 15. The prevalence rates of malaria, clinical malaria, and severe malaria were determined by calculating the number of cases per total number of subjects analysed. Descriptive statistics were used to determine averages and proportions. Participants with missing data were excluded from the analysis. Subgroup analyses for age, sex, and season were performed, and kappa analysis was used to control the quality of malaria diagnosis.
3 Results
3.1 Subjects
The main results of the study, including subject disposition and demographic information, were previously published [
6]. The overall malaria prevalence is reported here to present the context of malaria endemicity in the original study. A total of 1017 children were included in the study across the rainy and dry seasons.
3.2 Prevalence of malaria parasitaemia
The overall prevalence of malaria in the parent study was 34.8% as determined by microscopy. All parasites identified in this study were P. falciparum.
The prevalence rates across the rainy and dry seasons were similar for parasite and spleen rates as shown in Tab.1.
3.3 Prevalence of severe malaria across seasons
The overall prevalence of severe malaria was 2.6%, which was considerably lower in the dry season than in the rainy season (Tab.2). Different subjects had various criteria for severity.
3.4 Age distribution of severe malaria cases
Almost half of the children with severe malaria were 2 years of age with a general reduction in prevalence with age as shown in Fig.1. The mean age was 3.73 years (95% CI 2.81–4.64) with a standard error of 0.445.
3.5 Sex distribution of severe malaria cases
When categorised by sex, the ratio of male to female children with severe malaria was approximately 2:1 as shown in Tab.3. The sex ratio of the enrolled children was almost 1:1.
3.6 Parasite density across seasons
The mean parasite density was much higher during the rainy season (1006.13), and the standard deviation was 495.8 with minimum and maximum values of 840 and 284 000 parasites/µL, respectively. The dry season mean parasite density was lower at 405.45 parasites/µL, and the standard deviation was 209.48 with minimum and maximum values of 16 and 204 000 parasites/µL, respectively.
However, no remarkable differences were found when the parasite counts of the children with severe malaria during the rainy and dry seasons were plotted. The mean parasite density during the rainy season was much higher than that during the dry season. The small numbers disallow the determination of the significance of the difference between the means (Fig.2).
3.7 Level of parasitaemia among children with malaria
Of the 26 children with severe malaria, 14 (54.0%) had hyperparasitaemia as a criterion for the severity of malaria across the seasons. The prevalence of hyperparasitaemia in all study participants was 1.4%. The other children were classified as shown in Tab.4.
3.8 Pattern of severe malaria presentation
Most children had fever as a presenting feature of severe malaria (92.3%), with 73.0% presenting with vomiting. Other common features were multiple convulsions, lethargy, and altered sensorium. Other non-diagnostic features included severe clinical pallor, dehydration, and headache in older children. The clinical features of children with severe malaria are presented in Tab.5.
The criteria for the severity of malaria are at the top of the table with other symptoms. The children with a higher fever grade did not have the highest degree of parasitaemia, indicating that axillary temperature at diagnosis and parasite count had no relationship (Fig.3).
4 Discussion
The details of severe malaria cases in a community-based study designed to map the intensity of malaria transmission and the pattern of clinical disease in the LGA Wamakko, Sokoto State, Northwestern Nigeria are presented. Haemoglobin was not measured in the parent study because the aim was only to describe the parasite positivity rate as previously described [
6].
The overall PfPR2-10 was 34.8% [
8]. The area is mesoendemic for malaria transmission (intermediate endemicity). The implication of this classification on the pattern of severe malaria is that the mean age of children with severe malaria should be high with relatively high rates of cerebral malaria. As the intensity of malaria transmission decreases, the clinical pattern should shift towards cerebral malaria with older children being affected [
9]. In Uganda, the mean age of children with severe malaria decreased with increasing transmission intensity with an increased prevalence of severe anaemia, seizures, and coma [
10]. Conversely, in Kilifi, Kenya, the average age of children with severe malaria has increased in over 18 years, during which the intensity of transmission has been steadily declining, and the pattern of disease has shifted, favouring more cerebral malaria cases. The case fatality rate also increased likely because of the age shift and fatality associated with cerebral malaria [
11]. Hence, the mean age of children in our study with severe malaria is unexpectedly low at 3.73 years with approximately 50% of children with severe malaria aged 2 years. In addition, over the past 10 years in Sokoto, the endemicity of transmission has decreased as indicated by serial malaria indicator studies performed in 2010 [
12] and 2015 [
13].
Our results showed that the prevalence of non-anaemia severe malaria in this community-based study was 2.6%, which was much lower than the hospital-based study in Sokoto by Jiya
et al. [
14], who found a prevalence of 29.6% among the children admitted to the Emergency Paediatric Unit of the referral hospital in Sokoto, although they analysed the anaemia and non-anaemia forms of severe malaria. Another study by Olanrewaju and Johnson in a tertiary hospital in Ilorin found that a third of hospitalised children had severe malaria [
15]. The present results was also lower than the 3% prevalence of community-based hyperparasitaemia described by Olanrewaju
et al., who suggested that the prevalence of severe malaria is likely to be higher than 3% in their study, considering that children are likely to have other criteria for malaria severity. In our study, only 54.0% of children with severe malaria had hyperparasitaemia as a criterion for malaria severity. This value is much higher than the 17.1% reported by Jiya
et al., which was obtained in a different setting. The high percentage in our study means that these subjects would have been missed owing to the absence of facilities for malaria microscopy in the community setting. The higher proportion of hyperparasitaemia in our study was unexpected because higher parasite levels are expected in areas of higher transmission intensity [
10] and because the denominator is smaller, which in our case is a historic comparison. One subject with hyperparasitaemia did not present fever and other features of severe malaria, which contradicts the observation from a previous study that an increase in parasitaemia greater than 350 000 per microliter gave rise to a 1 °C rise in temperature above 37.5 °C [
16]. The most common presenting feature among children with severe malaria was fever, although not all children had fever, which can lead to the delayed diagnosis of severe malaria in many cases, particularly in the community.
The mean age of children with severe malaria in our study was 3.7 years, which was higher than that (3.5 years) reported previously in the same location [
14]. This finding is expected based on the reduction in transmission intensity over the years. The proportion of children with severe malaria who were 2 years of age in our study was 46%. This average age might be misleading, because only children aged 2–10 years were included in our study. This result suggests that the selected age range in this study likely excluded a number of children under 2 years as shown in MIS 2010 and 2015, which show a large proportion of children with malaria under 2 years of age [
13]. This age group might have had severe malaria in the community, and what we found might only be the tip of the iceberg. Therefore, community health providers and local medicine vendors need to be educated on the symptoms and signs of severe malaria to ensure prompt recognition and early referral to referral hospitals and reduce the mortality in the community from the disease, which often goes unrecorded and unaccounted for in the estimates of severe malaria [
2]. These vendors are usually the first point of call when children are ill [
17,
18] and play an important role in controlling severe malaria-related mortality. Another opportunity that could intervene in the community cases of severe malaria is seasonal malaria chemoprophylaxis, wherein field workers can be given a checklist for the diagnosis of severe malaria with rapid diagnostic testing for malaria when severe malaria is suspected. This measure can allow for the early referral of suspected cases of severe malaria and generate timely data on the community prevalence of this life-threatening condition.
The distribution of parasite density and the prevalence of severe malaria were higher in the rainy season than in the dry season and were intricately linked.
The sex distribution was strikingly in favour of males (2:1) despite the almost equal number of males and females included in the study. This finding is in line with similar reports of male predominance, despite the lack of a clear biological rationale for this finding, particularly considering the peak age of the children who spend their days around their mothers.
5 Conclusions
We observed a relatively high community-based prevalence of severe malaria, which is likely underestimated. The cases of severe malaria are well characterised, implying that interventions can be tailored to address this problem with a likelihood of success. Our data underline the unreliability of available malaria morbidity and mortality indices and suggest the need to continuously generate data to evolve disease control strategies.
The study was limited by the age range of children included in the study, which excluded children with severe malaria under 2 years of age; the use of only clinical criteria to define malaria severity; and the absence of data on laboratory results (e.g., hemoglobin) and severe malaria with anaemia because of the study design. The prevalence may have been higher if these factors were considered.
PDF
(576KB)
