Epidemiology of Plasmodium malariae and Plasmodium ovale spp. in Kinshasa Province, Democratic Republic of Congo

Abstract

Reports suggest non-falciparum species are an underappreciated cause of malaria in sub-Saharan Africa but their epidemiology is ill-defined, particularly in highly malaria-endemic regions. We estimated incidence and prevalence of PCR-confirmed non-falciparum and Plasmodium falciparum malaria infections within a longitudinal study conducted in Kinshasa, Democratic Republic of Congo (DRC) between 2015-2017. Children and adults were sampled at biannual household surveys and routine clinic visits. Among 9,089 samples from 1,565 participants, incidences of P. malariae, P. ovale spp., and P. falciparum infections by 1-year were 7.8% (95% CI: 6.4%-9.1%), 4.8% (95% CI: 3.7%-5.9%) and 57.5% (95% CI: 54.4%-60.5%), respectively. Non-falciparum prevalences were higher in school-age children, rural and peri-urban sites, and P. falciparum co-infections. P. falciparum remains the primary driver of malaria in the DRC, though non-falciparum species also pose an infection risk. As P. falciparum interventions gain traction in high-burden settings, continued surveillance and improved understanding of non-falciparum infections are warranted.

Fig. 1. Study population.

The study population comprised 1565 enrolled participants who had a DBS sample at the baseline visit. Participants without a baseline DBS were excluded from analyses. Participant loss-to-follow-up among the survey-based population is shown, with 76% completing all three follow-up visits. Sixty-seven percent of participants in the survey population visited a study health clinic at least once during follow-up, comprising the clinic-based subpopulation. Study health clinic visit frequency across the full study period is shown. The total population encompassed all survey and clinic-based study visits. DBS dried blood spot, FU follow-up.

Fig. 2. Cumulative incidence of Plasmodium infections across 34-months of follow-up in the total study population.

a Cumulative incidence curves depict time to the first detected P. malariae (blue), P. ovale (red), and P. falciparum (yellow) infections among participants negative for each Plasmodium species infection at baseline. Shading depicts 95% CIs around cumulative incidence estimates. Incidence curves increase in a step-wise fashion at follow-up survey timepoints due to participant-wide screening and case detection, whereas gradual slopes in between survey timepoints indicate additional incident events detected through symptomatic presentation to clinics. b Crude risks of incident infections among the total population shown in Fig. 2a, by malaria species. Time to the first incident infection since baseline. CI confidence interval, Pf Plasmodium falciparum, Pm Plasmodium malariae, Po Plasmodium ovale spp.

Fig. 3. Cumulative incidence curves stratified by participant characteristics.

Cumulative incidence of first infection stratified by age at baseline, for P. malariae (n = 1518; <5 years: n = 296, 5–14 years: n = 470; 15+ years: n = 752 at risk at baseline) in blue (a), P. ovale spp. (n = 1557; <5 years: n = 301, 5–14 years: n = 498; 15+ years: n = 758 at risk at baseline) in red (b), and P. falciparum (n = 1081; <5 years: n = 238, 5–14 years: n = 260; 15+ years: n = 583 at risk at baseline) in yellow (c). Cumulative incidence of first infection stratified by sex, for P. malariae (n = 1518; female: n = 840, male: n = 678 at risk at baseline) in blue (d), P. ovale spp. (n = 1559; female: n = 858, male: n = 701 at risk at baseline) in red (e), and P. falciparum (n = 1081; female: n = 618 male: n = 463 at risk at baseline) in yellow (f). p-values are calculated from log-rank testing. Shading depicts 95% CIs around incidence estimates. CI confidence interval.

Fig. 4. Factors associated with P. malariae and P. ovale spp. infection prevalence, stratified by survey vs. clinic-based population.

a Factors associated with P. malariae (blue) and P. ovale (red) infection prevalences at survey visits (baseline and three follow-up surveys). b Factors associated with P. malariae (blue) and P. ovale (red) prevalences at symptomatic clinic visits across follow-up (i.e., symptomatic cases). Error bars represent 95% CIs around unadjusted prevalence difference measures; numerical values of prevalence differences and 95% CIs are listed in Supplementary Tables 8 and 9. n = 5659 biologically independent samples in the survey population (a), and n = 3349 in the clinic-based population (b). Factors associated with non-falciparum infection prevalences compared to P. falciparum infection are depicted in Supplemental Fig. 2. CI confidence interval, Pf Plasmodium falciparum, Pm Plasmodium malariae, Po Plasmodium ovale spp., RDT rapid diagnostic test.

Fig. 5. Distributions of P. falciparum parasitemia by type of non-falciparum mixed infection vs. P. falciparum mono infections, stratified by population.

Household survey-based results for P. falciparum parasitemia distributions comparing P. falciparum mono infections (n = 3418) in yellow vs. co-infection with (a) P. malariae (n = 206) in blue and (b) P. ovale spp. (n = 96) in red. Clinic sub-population P. falciparum parasitemia distributions comparing P. falciparum mono infections (n = 1834) in yellow vs. co-infection with (c) P. malariae (n = 84) in blue and (d) P. ovale spp. (n = 47) in red. Density distributions are displayed to illustrate trends in overall distributions accounting for variation in sample sizes across groups. Pf Plasmodium falciparum.

Fig. 6. Study design overview.

Biannual household study visits occurred at baseline and three follow-ups over a 20-month longitudinal period in which household demographics and individual health surveys were collected, malaria RDTs were performed, and DBS were collected. Participants who developed fever or other malaria symptoms during follow-up visited study health clinics, where clinical questionnaires, malaria RDTs, a DBS, and hemoglobin testing were performed. Study participants completing household surveys were included in the survey population; those who also visited the clinic at least once for malaria symptoms were included in the clinic subpopulation. DBS dried blood spot, RDT rapid diagnostic test.