Comparison of Plasmodium ovale curtisi and Plasmodium ovale wallikeri infections by a meta-analysis approach

Abstract

Malaria caused by Plasmodium ovale species is considered a neglected tropical disease with limited information about its characteristics. It also remains unclear whether the two distinct species P. ovale curtisi and P. ovale wallikeri exhibit differences in their prevalence, geographic distribution, clinical characteristics, or laboratory parameters. Therefore, this study was conducted to clarify these differences to support global malaria control and eradication programs. Studies reporting the occurrence of P. ovale curtisi and P. ovale wallikeri were explored in databases. Differences in proportion, clinical data, and laboratory parameters between the two species were estimated using a random-effects model and expressed as pooled odds ratios (ORs), mean difference (MD), or standardized MD depending on the types of extracted data. The difference in geographical distribution was visualized by mapping the origin of the two species. A total of 1453 P. ovale cases extracted from 35 studies were included in the meta-analysis. The p-value in the meta-analyses provided evidence favoring a real difference between P. ovale curtisi malaria cases (809/1453, 55.7%) and P. ovale wallikeri malaria cases (644/1453, 44.3%) (p: 0.01, OR 1.61, 95% CI 0.71-3.63, I²: 77%). Subgroup analyses established evidence favoring a real difference between P. ovale curtisi and P. ovale wallikeri malaria cases among the imported cases (p: 0.02, 1135 cases). The p value in the meta-analyses provided evidence favoring a real difference in the mean latency period between P. ovale curtisi (289 cases) and P. ovale wallikeri malaria (266 cases) (p: 0.03, MD: 27.59, 95% CI 1.99-53.2, I²: 94%), total leukocyte count (p < 0.0001, MD: 840, 95% CI 610-1070, I²: 0%, two studies) and platelet count (p < 0.0001, MD: 44,750, 95% CI 2900-60,500, I²: 32%, three studies). Four continents were found to have reports of P. ovale spp., among which Africa had the highest number of reports for both P. ovale spp. in its 37 countries, with a global proportion of 94.46%, and an almost equal distribution of both P. ovale spp., where P. ovale curtisi and P. ovale wallikeri reflected 53.09% and 46.90% of the continent's proportion, respectively. This is the first systematic review and meta-analysis to demonstrate the differences in the characteristics of the two distinct P. ovale species. Malaria caused by P. ovale curtisi was found in higher proportions among imported cases and had longer latency periods, higher platelet counts, and higher total leukocyte counts than malaria caused by P. ovale wallikeri. Further studies with a larger sample size are required to confirm the differences or similarities between these two species to promote malaria control and effective eradication programs.

Figure 1

The study flow diagram.

Figure 2

The geographical distribution of P. ovale spp. The map was generated by authors using the map freely available at https://mapchart.net/. Authors are allowed to use, edit and modify any map created with mapchart.net for publication freely by adding the reference to mapchart.net. The project of is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Figure 3

The pooled proportion of P. ovale curtisi malaria. ES estimated proportion, CI confidence interval, Random random-effects model. The study conducted by Gabrielli et al. (2016) was excluded from the analysis because no P. ovale curtisi case was reported.

Figure 4

The pooled proportion of P. ovale wallikeri malaria. ES estimated proportion, CI confidence interval, Random random-effects model. The studies conducted by Haiyambo et al. (2018), Li et al. (2016), Díaz et al. (2015), and Krishna et al. (2017) were excluded from the analysis because no P. ovale curtisi case was reported.

Figure 5

The difference in proportion between P. ovale curtisi and P. ovale wallikeri malaria among imported and indigenous cases. IV inverse variance, CI confidence interval, Event Number of P. ovale curtisi or P. ovale wallikeri cases, Random random-effects model, Total number of all P. ovale spp. cases, Lower in curtisi: The proportion of P. ovale curtisi cases was lower than that of P. ovale wallikeri cases. Higher in curtisi: The proportion of P. ovale curtisi cases was higher than that of P. ovale wallikeri cases.

Figure 6

The difference in age between P. ovale curtisi and P. ovale wallikeri malaria. IV inverse variance, CI confidence interval, SD standard deviation, Random random-effects model, Total number of all P. ovale spp. cases, Lower in curtisi: The mean age of P. ovale curtisi cases was lower than that of P. ovale wallikeri cases. Higher in curtisi: The mean age of P. ovale curtisi cases was higher than that of P. ovale wallikeri cases.

Figure 7

The difference in gender between P. ovale curtisi and P. ovale wallikeri malaria. IV inverse variance, CI confidence interval, Event number of male patients with P. ovale curtisi or P. ovale wallikeri, Random random-effects model, Total number of all P. ovale spp. cases, Lower in curtisi: The proportion of male patients with P. ovale curtisi cases was lower than that of P. ovale wallikeri cases. Higher in curtisi: The proportion of male patients with P. ovale curtisi cases was higher than that of P. ovale wallikeri cases.

Figure 8

The difference in latency period between P. ovale curtisi and P. ovale wallikeri malaria. IV inverse variance, CI confidence interval, SD standard deviation, Random random-effects model, Total number of all P. ovale spp. cases, Shorter in curtisi: The mean latency period of P. ovale curtisi cases was shorter than that of P. ovale wallikeri cases. Longer in curtisi: The mean latency period of P. ovale curtisi cases was longer than that of P. ovale wallikeri cases.

Figure 9

The difference in parasite density between P. ovale curtisi and P. ovale wallikeri malaria. IV inverse variance, CI confidence interval, SD standard deviation, Std. mean difference Standard mean difference (SMD), Random random-effects model, Total number of all P. ovale spp. cases, Lower in curtisi: The SMD of parasite density in P. ovale curtisi cases was lower than that in P. ovale wallikeri cases. Higher in curtisi: The SMD of parasite density in P. ovale curtisi cases was higher than that in P. ovale wallikeri cases. Mean of parasite density is in parasites per microliter (parasites/µL).

Figure 10

The difference in hemoglobin level between P. ovale curtisi and P. ovale wallikeri malaria. IV inverse variance, CI confidence interval, SD standard deviation, Random random-effects model, Total number of all P. ovale spp. cases, Lower in curtisi: The mean hemoglobin of P. ovale curtisi cases was lower than that of P. ovale wallikeri cases. Higher in curtisi: The mean hemoglobin level of P. ovale curtisi cases was higher than that of P. ovale wallikeri cases.

Figure 11

The difference in leukocyte count between P. ovale curtisi and P. ovale wallikeri malaria. IV inverse variance, CI confidence interval, SD standard deviation, Random random-effects model, Total number of all P. ovale spp. cases, Lower in curtisi: The mean leukocyte counts of P. ovale curtisi cases was lower than that of P. ovale wallikeri cases. Higher in curtisi: The mean leukocyte counts of P. ovale curtisi cases were higher than those of P. ovale wallikeri cases.

Figure 12

The difference in platelet counts between P. ovale curtisi and P. ovale wallikeri malaria. IV inverse variance, CI confidence interval, SD standard deviation, Random random-effects model, Total number of all P. ovale spp. cases, Lower in curtisi: The mean platelet counts of P. ovale curtisi cases were lower than those of P. ovale wallikeri cases. Higher in curtisi: The mean platelet counts of P. ovale curtisi cases were higher than those of P. ovale wallikeri cases.

Figure 13

Funnel plot. SE standard error, OR odds ratio.