Asymptomatic and sub-microscopic Plasmodium falciparum infection in children in the Mount Cameroon area: a cross-sectional study on altitudinal influence, haematological parameters and risk factors

The study was carried out in Batoke (Limbe), Dibanda (Buea) and Tole (Buea), which are three semi-rural communities along different altitudinal ranges in the Mount Cameroon area, as shown in Fig. 1. The sites were classified as lowlands (< 200 m above sea level (asl)), middle belt (200–600 m asl) and highlands (> 600 m asl). The coordinates of Batoke range from altitude 8 m asl, latitude 04°01.364′ N, longitude 009°05.971′ E to 47 m, 04°02.039′ N and 009°05.808′ E; Dibanda from altitude 358 m asl, latitude 04°06.447′ N, longitude 009°18.725′ E to 400 m asl, 04°07.179′ N and 009°18.464′ E and Tole is located between 627 m asl, latitude 04°07.057′N, longitude 009°15.178′E and 630 m asl, latitude 04°6.906′N, longitude 09°14.434′E. These three study sites have been described in detail by Teh et al. [3, 23].

The work was conducted among pre- and school aged children of both sexes between 6 months and 14 years old, who granted assent and whose parents/caregivers consented to participate in the study. The inclusion and exclusion criteria of this study were revised according to Teh et al. [3].

This cross-sectional community-based study was conducted in the rainy season between the months of July and October 2017 and between April and May 2018 in Tole, Batoke and Dibanda. Participants were invited for data collection in each community by their local chiefs and coordination was organised by the head/leader of a block within a neighbourhood (“quarter head”) of the various communities. Potential participants were reminded of the collection dates per block by the head/leader of the block. At the start of the study in each site, the parents, guardians and children were educated on the study protocol and the benefits of participation highlighted at their various neighbourhoods using an information sheet. Upon obtaining informed consent/assent from the parents/care givers, the study team proceeded with the collection of samples at specific identified collect sites.

The sample size for each study altitude was calculated using the 66.2% prevalence of malaria in children in the study area [24]. Sample size was determined using the formula n = Z²pq/d² [25] where n = the sample size required, z = 1.96: which is the standard normal deviate (for a 95% confidence interval, CI), p = 66.2%: proportion of malaria prevalence, q = 1—p: proportion of malaria negative children and d = acceptable error willing to be committed. The minimum sample size was estimated as n = 344 for each site. Considering a possible participation of more than one child per family, loss of samples due to blood clotting and incomplete data entry, the sample size was adjusted by 10% to a minimum of 379.

A multistage cluster sampling method was used to obtain the required sample. In the first stage, all the communities in the Mount Cameroon area were stratified into 3 zones namely lowland, Middlebelt and highland. One community was randomly selected from each zone namely, Tole (highland), Dibanda (middle belt) and Batoke (lowland). In the second stage, 21 clusters (quarters) were randomly selected from 31 clusters within the three communities. Within the clusters, all the households with children ≤ 14 years of age were selected. In a household where only one child within that age was present, the child was selected automatically. In a case where more than one child was present in a household, only one was randomly selected. A probability proportionate to size sampling method was used to select 400 random malaria negative samples from all the negative samples in the study population for the sub-microscopic studies.

Sociodemographic data which included information on gender, age, literacy level of parents/caregivers and malaria preventive methods, as well as fever history of participants were collected using a structured questionnaire. Axillary temperature was measured using an electronic thermometer with a febrile condition considered as temperature ≥ 37.5 °C [26].

Three to four (3–4) millilitres (mL) of venous blood sample was collected from each child using sterile disposable syringes. Part of the blood sample was used to prepare thick and thin films on the same slide for the determination of the presence of malaria parasite by Giemsa-stain microscopy using standard methods [27]. Parasite densities were expressed as asexual parasites per µL with reference to the participant’s white blood cell count (WBC) and categorized as low, moderate, high and hyper parasitaemia [28].

Also, 50 μL of the EDTA blood sample was aliquoted onto a Whatman 3 mm filter paper and dried overnight at room temperature. The dried blood spots (DBS) were used to determine sub-microscopic Plasmodium infection. Genomic deoxyribonucleic acid (DNA) was isolated from the DBS using chelex [29]. Primary and nested polymerase chain reaction (PCR) assays were carried out for all genes. The primary PCR was carried out with a pair of Plasmodium genus-specific primers (rPLU5-5ʹCCTGTTGTTGCCTTAAACTTC3ʹ and rPLUS6-5ʹTTAAAATTGTTGCAGTTAAAACG3ʹ) which amplified a 1100-base pair (bp) PCR product from the rRNA small subunit gene (18S rRNA) while the nested primers specific for P. falciparum (rFAL1-5ʹTTAAACTGGTTTGGGAAAACCAAATATATT3ʹ) and rFal2-5ʹACACAATGAACTCAATCATGACTACCCGTC3ʹ) were used, which amplified a 205-bp indicating a P. falciparum infection [30].

Furthermore, an auto-haematology analyser (Urit-3300® analyser, Guangxi, China) was used to assess haematological parameters following the manufacturer’s instructions and the condition of anaemia was defined as haemoglobin level (Hb) < 11 g/dL of whole blood [27].

Continuous variables were summarized into means and standard deviations (SD) and categorical variables reported as frequencies and percentages, were used to evaluate the descriptive statistics. The differences in proportions were evaluated using Pearson’s Chi-Square (χ²). Group means were compared using Kruskal Wallis and Mann–Whitney U Test. Parasite densities were log transformed before analysis. Associations between predictor variables and primary outcomes were assessed using both bivariate and multivariate logistic regression analysis. Odd ratios (ORs) and 95% confidence intervals (CIs) were computed. Any covariate with a P value < 0.2 in the bivariate analysis was subsequently included in the final multivariable logistic model. Significant levels were measured at 95%CI with the level of significance set at P < 0.05. Post entry and clean-up of data in Microsoft Excel 2016, analysis was performed using the IBM-Statistical Package for Social Sciences (IBM-SPSS) version 20 and Epi-info version 7.

A total of 1319 children with a mean (SD) age of 6.0 (3.5) years, residing at lowland (30.7%, 405), middle belt (37.2%, 491) and highland (32.1%, 423) in the Mount Cameroon area were evaluated. As shown in Table 1, most of the parents/caregiver of the children had a primary (47.9%) and secondary (31.6%) level of education. The proportion of febrile children in the study population was 8.5% (112), with no significant differences in age and sex. The prevalence of malaria parasite, asymptomatic malaria parasitaemia and sub-microscopic infection in the study population was 36.4%, 34.0% and 43.8%, respectively. Children between 5 and 9 years had the highest occurrence of microscopic (39.4%) and asymptomatic parasitaemia (36.9%) at P = 0.021 and P = 0.047 respectively, when compared with their contemporaries.

The prevalence of sub microscopic malaria was significantly highest (P < 0.001) among the 10–14 years age group (60.5%) when compared with the other age groups (Table 1).

The overall microscopic prevalence of malaria parasite was 36.4% (CI: 33.8–39.0). The prevalence of asymptomatic falciparum malaria among the 1271 children without fever varied with altitude. The overall prevalence in the low, middle belt and high lands was 44.6% (CI: 39.8–49.6), 25.2% (CI: 39.8–49) and 34.1% (CI: 29.6–38.8), respectively, and the difference was statistically significant (χ² = 35.980, P < 0.001) (Fig. 2). In the lowland, asymptomatic malaria was significantly highest (χ² = 6.651, P = 0.036) in children aged 5‒9 years (52.2%, CI: 44.5–59.9) old while, in the middle belt it was significantly highest (χ² = 7.007, P = 0.03) in children < 5 years (29.5%, CI: 23.7–36.0) when compared with their respective counterparts. No significant difference with age was observed in highlands even though the prevalence was highest in children 5–9 years old (37.4%, CI: 30.5–44.9). On the other hand, while children < 5 years in the low, middle belt and high lands had similar prevalence of asymptomatic malaria parasitaemia, those 5–9 years (52.2%, CI: 44.5–59.9) and 10–14 years (39.5%, CI: 29.9–50.1) resident in the lowlands, had the highest prevalence and the difference was significant (χ² = 28.830, P < 0.001 and χ² = 12.720, P = 0.002, respectively). As shown in Fig. 2, asymptomatic malaria prevalence among the sexes was comparable within the low and high lands but statistically different in the middle belt (χ² = 5.157, P = 0.023) where, females had higher prevalence (29.3%, CI: 24.1–35.0) than males (20.2%, CI: 15.4–26.1). Conversely, significantly higher (χ² = 32.251, P < 0.001 and χ² = 8.896, P = 0.012) prevalence was observed in males (46.5%, CI: 39.7–53.4) and females (42.6%, CI: 35.8–49.7) in the lowland when compared with those in the middle belt and highland, respectively.

The geometric mean parasite density (GMPD) was significantly higher (P < 0.001) in children residing in the lowland (449 parasites/ µL of blood) those < 5 years old in the lowland (538 parasites/ µL of blood, P = 0.024) and middle belt (224 parasites/ µL of blood, P = 0.003) and males (218 parasites/ µL of blood, P = 0.025) in the middle belt (see Additional file 1).

The prevalence of low, moderate and high parasitaemia in the study population were 84.2% (401/476), 12.4% (59/476) and 3.4% (16/476). As shown in Fig. 3, the prevalence of low, moderate and high malaria parasitaemia was significantly different (P < 0.001) in children from the different altitudes, with the low parasite density category being the most prevalent in all the three settings.

Overall, the prevalence of sub-microscopic malaria parasitaemia with respect to altitude was highest in children in the highland (66.7%) and lowest in the lowland (29.2%). With respect to age related differences and altitude, among children < 5 years, those resident in middle belt had the highest sub-microscopic malaria parasite prevalence (34.1%). On the other hand, the 5–9 and 10–14 years old, resident in highland had significantly higher (P < 0.001) prevalence (87.2% and 81.8%, respectively) than those in the other altitudinal sites. Moreover, males and female’s resident in the highlands had the highest prevalence of sub-microscopic infection (75.4% and 58.7%), compared to the other altitudinal sites at P < 0.001 and P = 0.002, respectively, as shown in Table 2.

The overall prevalence of anaemia was 62.3%. No significant differences were observed with sex and febrile status while, the prevalence of anaemia decreased significantly (P < 0.001) with an increase in age with youngest age group having a prevalence of 73.8%. The occurrence of anaemia was significantly higher in children from the middle belt (72.7%), those with asymptomatic microscopic (68.1%) and sub-microscopic (78.3%) Plasmodium infection than their respective equal (Table 3). Relating to the severity of anaemia, children aged < 5 years had the highest prevalence of severe (7.8%) and moderate (58.8%) anaemia while mild anaemia was most common in those 10‒14 years old (53.4%) and the difference was statistically significant at P < 0.001. Significantly (P < 0.001 and P = 0.006), moderate anaemia was the most occurring form of anaemia in children negative for asymptomatic malaria parasitaemia (56.7%) and those positive for sub-microscopic infection (64.2%), respectively, as shown in Table 3.

The mean haematological parameters were comparable between children with and without sub microscopic infection except for the mean Hb levels, haematocrit (Hct), RBC (red blood cell) and platelet (Plt) counts, mean corpuscular haemoglobin concentration (MCHC) and red cell distribution–coefficient of variation (RDW-CV). Children with sub microscopic infection had a significantly lower mean Hb concentration (9.86 ± 1.7 g/dL), RBC (4.48 ± 1.1 × 10¹²/L) and Plt (280.83 ± 112.62) counts, Hct (31.92%) and MCHC (31.33 ± 4.74 g/L) than their negative counterparts as shown in Table 4. On the other hand, the mean RDW-CV% (15.19 ± 3.3) was significantly higher (P < 0.001) in children with sub-microscopic infection than those negative.

The logistic regression model with sub-microscopic infection status as dependent variable and altitude, age, gender, marital status, ITN usage, fever, fever within a month, anaemia and water source as independent variable, demonstrated that children from highlands (P =  < 0.001), those between 5 and 9 years (P =  < 0.001) and 10‒14 years (P =  < 0.001), those who did not use ITN (P = 0.04) and anaemic children (P =  < 0.001) were more likely to have sub-microscopic malaria parasite infection. The odds of carrying sub-microscopic infection is presented in Table 5. Children from highlands, those 5‒9 years and between 10 and 14 years of age, who did not use ITN and anaemic as well were 1.8, 3, 8, 1.69 and 9 times greater the odds to carry sub-microscopic Plasmodium infection than their counterparts.