A few studies have examined the effect of asymptomatic parasitaemia on cognition (Al Serouri et al. 2000; Boivin et al. 1993; Vitor-Silva et al. 2009), whilst others have determined cognitive outcomes in children using malaria prophylaxis to prevent the development of parasitaemia (Clarke et al. 2008; Fernando et al. 2006; Jukes et al. 2006; Thuilliez et al. 2010).

In the Yemeni study, 445 boys with asymptomatic parasitaemic boys were compared with 142 non-parasitaemic boys matched for age and schooling status (Al Serouri et al. 2000). Two weeks later, 150 children who remained parasitaemic and 150 children who were no longer parasitaemic were assessed using a battery of cognitive tests. The investigators found no significant differences in cognitive test scores between those who became non-parasitaemic and those who remained parasitaemic, although there was a difference in fine motor tasks.

Boivin et al. (1993) evaluated 97 Zairian schoolchildren for cognitive ability. Even though treating asymptomatic malaria did not improve cognitive ability, clearance of intestinal parasites showed improved performance on the Kaufman Assessment Battery for children.

A study of 198 Brazilian children aged 5–14 years sought to determine the cognitive outcomes of malaria parasite elimination for a period of 9 months (Vitor-Silva et al. 2009). Poor performance of language and mathematics was associated with at least one malarial attack. Overall, it remains unknown whether parasitaemia on its own results in poorer cognitive function or factors are involved.

Fernando (2001) reported that the academic performance of 293 children experiencing uncomplicated malaria was poorer than that of 162 children experiencing non-malarial fever or that of 305 healthy controls. Although the performance of the children with malaria improved 2 weeks after treatment, it was still significantly lower than that of the healthy controls. This study was unable to examine whether poor performance was due to malaria per se, or the absenteeism from school caused by having an illness.

The same investigators also conducted a randomized double-blind placebo controlled study of antimalarial prophylaxis with school-age Sri Lankan children, in which they measured academic performance using school tests of language and mathematics (Fernando 2001; Fernando et al. 2006). Two hundred and ninety-five children received chloroquine while another 292 children received placebo. They found an improvement in academic performance in those on prophylaxis for a 9-month period. However, the placebo group also had significantly higher rates of absenteeism. A multivariate model identified absenteeism due to malaria and chloroquine prophylaxis as significant predictors of school performance.

Similar studies investigating the effect of intermittent preventive treatment (IPT) on education outcomes have been carried out. Clarke et al. (2008) and colleagues studied children aged 5–18 years who received three treatments at 4-month intervals and outcome was assessed through cross-sectional surveys 12 months post-intervention. Children in the intervention group performed significantly better on the sustained attention task but differences were shown in education achievement. Jukes et al. (2006) followed up children in the Gambia who were part of randomized trial done 15 years earlier. The study examined the long-term impact of malaria prophylaxis on cognitive abilities and found no overall intervention effect on cognitive abilities even though the intervention group had higher educational attainment. A prospective study of 227 Malian primary school children, in which each child had monthly clinical and lab examinations and tests for cognitive ability found poorer outcomes in the children with asymptomatic malaria, though which sub-tests showed significant differences was not reported (Thuilliez et al. 2010).

There are a number of studies investigating the effects of severe malaria on cognitive function following exposure to severe malaria, (Bangirana et al. 2011; Boivin et al. 2007, 2011; Carter et al. 2003, 2005; Holding et al. 1999, 2004; John et al. 2008). Severe malaria manifests as impaired consciousness, prostration, seizures, severe anaemia, acidosis, hypoglycaemia or renal impairment. Impaired consciousness is defined as Blantyre Coma Score (BCS) of ≤4 for 4 or more hours. Cerebral malaria is the most severe neurological manifestation of falciparum malaria (Newton et al. 2000). The World Health Organization (2000) defines it as one’s inability to localise a painful stimulus or a Blantyre coma score of ≤2 (Molyneux et al. 1989), a peripheral falciparum malaria parasitaemia and exclusion of other causes of encephalopathy. More recently features of retinal changes were associated with the sequestration of the parasites within the brain and define a more specific syndrome of cerebral malaria (White et al. 2009).

Most studies have revealed poorer performance in children exposed to severe malaria except one small Gambian study of children who survived without neurological deficits and showed no differences compared to community controls (Muntendam et al. 1996).

In Kilifi, Kenya, Holding et al. (1999) compared 87 children with a history of severe malaria with impaired consciousness (defined as a Blantyre score of ≤4) with community controls on tasks measuring information processing, language and behaviour based on the K-ABC (Kaufman and Kaufman 1983) and other locally-established tests. There was no significant impairment of information processing skills for the children, although a greater number of cases showed impaired performance. There were significant deficits in measures of language and attention/planning.

Holding et al. (2004) represented a re-analysis of their previously reported data, in which children with a history of severe malaria were divided into medium- and high-risk groups based on the severity of the disease. Seventeen children were considered high-risk and the rest, medium-risk. These children were compared to controls (low-risk group) matched for age, sex and socioeconomic status. The results revealed the adverse effects of schooling on the outcome of disease. Unschooled high-risk children performed significantly poorer than unschooled children in the other two groups.

Carter et al. (2003) compared 25 children previously admitted to hospital with cerebral malaria (CM) or severe malaria (defined as malaria prostration, multiple seizures or severe anaemia) and 27 children unexposed to either condition and found that language performance was poorer in the children exposed to either CM or severe malaria up to 6 years post-insult. Assessments of comprehension, syntax, lexical semantics, higher-level language abilities, pragmatics and phonology were administered to each child. Children with a history of CM or severe malaria performed significantly poorer on tests of comprehension, syntax and lexical semantics compared to the unexposed group. There was no evidence of a difference between the scores of children who had suffered CM and those who had severe malaria, although this may have been due to the small numbers.

In another study (Carter et al. 2005), 156 children exposed to malaria with complicated seizures (defined as >2 seizures within 24 h or focal or prolonged seizures for >30 min but without coma) showed significantly increased odds of impairment with respect to speech and language (pragmatics and phonology) compared to unexposed children. There was no evidence of a difference in performance on tests of memory, attention or other aspects of language.

Many studies have reported cognitive deficits associated with severe malaria, particularly CM (Table 1). In a Ghanaian study, 20 children (aged 7–16 years) with a history of CM (defined using the WHO criteria) were compared with 20 age-, sex- and education-matched controls on a standardised neuropsychological battery (Dugbartey et al. 1998). Children with a history of CM performed significantly poorly relative to controls in bimanual tactile discrimination, accuracy of visual scanning, visual memory, perceptual abstraction and rule learning skill, right ear auditory information processing, and dominant-hand motor speed. The study found no significant differences between those with a history of CM and controls in non-verbal reasoning, visual-spatial processing, auditory attention and verbal fluency.

In Senegal, 29 children aged between 5 and 12 years with a history of CM (defined using WHO criteria but with coma duration adjusted to 12 h) were compared with 29 age- and education-matched controls (mild-malaria). Those with a history of CM performed significantly poorer on the simultaneous processing (spatial memory, photo series), mental processing, and sequential processing (hand movements, word order) tasks of the K-ABC and the attention capacity task from Test of Attention Variables (TOVA) than matched controls (Boivin 2002). The study found significant correlation between coma duration and attention capacity for the CM group emphasizing the importance of coma duration on outcome.

In Kenya, 152 children aged between 6 and 9 years previously exposed to CM were compared to 179 children unexposed to severe malaria (Carter et al. 2005). The performance of children previously exposed to CM was poorer than unexposed children on all the cognitive assessments administered: speech and language battery, attention, memory, and non-verbal functioning. Significant differences were found on tests of higher-level language abilities, lexical semantics, pragmatics and non-verbal functioning (construction task).

There was one study that did not detect any differences between CM and controls. This case-control study focused on the measurement of non-verbal functioning in 36 pairs of Gambian children who were discharged from hospital without any neurological deficits (Newton et al. 2000). The results showed no significant impairments in children after an average follow-up of 3.4 years (Muntendam et al. 1996). A curious finding from the study, though not statistically significant, was that the CM children performed better on all tests of intellectual development than the controls. A group difference of borderline significance was detected on a test of balance (sensori-motor development), with poorer performance among children exposed to CM.

A Ugandan study of children aged 5–12 years presenting with either cerebral malaria (n = 44), or uncomplicated malaria (n = 54) and compared with healthy community children (n = 89) found a 3.67 increased risk of cognitive impairment in CM children compared to controls (Boivin et al. 2007; John et al. 2008). At 6 months post discharge, the deficits were in working memory and attention (Boivin et al. 2007). Deficits in these groups persisted 2 years post discharge especially in areas of attention (John et al. 2008).

Boivin et al. (2011) assessed Malawian children with retinopathy-positive CM for the domains of gross motor, fine motor, language and social skills. Eighty-three CM children were compared with 95 community controls. The results of cognitive performance showed poorer language development in those exposed to CM. A Stepwise regression demonstrated that coma duration and seizures while in hospital were predictive of poor outcome.

Finally, a prospective case-control study done in Uganda of 62 children with a history of malaria and neurological involvement and 61 community controls, assessed working memory, reasoning, learning, visual spatial skills and attention 3 months post-discharge (Bangirana et al. 2011). The results showed marginal non-significant evidence of lower attention scores and differences in the other cognitive abilities. The authors argued that lack of differences between the malaria and control groups could be as a result of gradual development of cognitive deficits such that the effect is less obvious in the short term.