Severity and mortality of severe Plasmodium ovale infection: A systematic review and meta-analysis

Plasmodium ovale can infect humans, causing malaria disease. We aimed to investigate the severity and mortality of severe P. ovale infection to increase the awareness of physicians regarding the prognosis of this severe disease and outcome-related deaths in countries in which this disease is endemic. Articles that were published in the PubMed, Scopus, and ISI Web of Science databases prior to January 5, 2020 and reported the prevalence of severe P. ovale infection were systematically searched and reviewed. Studies that mainly reported severe P. ovale infection according to the 2014 WHO criteria for the treatment of malaria were included. Two reviewers selected, identified, assessed, and extracted data from studies independently. The pooled prevalence of severe P. ovale mono-infections was estimated using the command “metaprop case population, random/fixed”, which yielded the pooled estimate, 95% confidence interval (CI) and the I2 value, indicating the level of heterogeneity. Meta-analyses of the proportions were performed using a random-effects model to explore the different proportions of severity between patients with P. ovale and those with other Plasmodium species infections. Among the eight studies that were included and had a total of 1,365 ovale malaria cases, the pooled prevalence of severe P. ovale was 0.03 (95% CI = 0.03–0.05%, I2 = 54.4%). Jaundice (1.1%), severe anemia (0.88%), and pulmonary impairments (0.59%) were the most common severe complications found in patients infected with P. ovale. The meta-analysis demonstrated that a smaller proportion of patients with P. ovale than of patients with P. falciparum had severe infections (P-value = 0.01, OR = 0.36, 95% CI = 0.16–0.81, I2 = 72%). The mortality rate of severe P. ovale infections was 0.15% (2/1,365 cases). Although severe complications of P. ovale infections in patients are rare, it is very important to increase the awareness of physicians regarding the prognosis of severe P. ovale infections in patients, especially in a high-risk population.


Study selection
This systematic review was designed on the basis of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (S1 Checklist). Two authors (MK and KUK) searched the Medline, Scopus, and ISI Web of Science databases independently for articles published in English prior to January 5, 2020. Additional articles from other databases were also selected. The search terms, which included the Boolean operators "OR" or "AND", were as follows: "(severe OR complicated OR complication) AND (Plasmodium ovale)" (S1 Table). EndNote software version X7 (Thomson Reuters, New York, NY) was used to process all references in our study.

Quality of the included studies
The quality of the observational studies was assessed in accordance with the Newcastle-Ottawa Scale (NOS) [31]. A 'star system', in which a study is judged on the basis of the following three broad perspectives was used: the selection of the study groups; the comparability of the groups; and the ascertainment of either the exposure or outcome of interest.

Data extraction
All raw and available data from eight studies were extracted to explore the prevalence of severe complications in patients with P. ovale infections. Two authors (MK and KUK) extracted the data from the selected studies. If there were any discrepancies between the two reviewers, another reviewer (GM) determined whether the study should be included. For the prevalence outcome of severe P. ovale malaria, all prospective cohort, cross-sectional studies, and casecontrol studies that reported the number of patients with severe complications from P. ovale were included. The following articles were excluded from our study: book and book chapter reviews, conference papers, editorials, letters, correspondences, notes, reviews, animal studies, case reports, drug studies and clinical trials, entomological studies, experimental studies, knowledge assessments, studies written in local languages, studies on mixed infections of P. ovale and other Plasmodium spp., studies of P. ovale during pregnancy, prevalence studies of non-P. ovale infections, and prevalence studies of P. ovale that did not report data on severe complications.

Statistical analysis
The primary outcome of the present study was the pooled prevalence of severe P. ovale infections and mortality rates. The number of severe P. ovale infections and the total number of P. ovale infections were entered into an Excel data sheet (Microsoft Corporation, USA). The pooled prevalence of the severity of P. ovale mono-infections was estimated using the command "metaprop case population, random/fixed" available in STATA software version 15.0 (StataCorp LLC, USA). The results are presented as the pooled estimate and 95% confidence interval (CI). A meta-regression with the median age as a covariate was performed to evaluate whether the age of the participants was a confounder of the pooled prevalence of severe P. ovale.
The secondary outcome of the present study was the different proportions of severity between patients with P. ovale and those with other Plasmodium species. The meta-analyses of the proportions were performed using a random-effects model provided in Review Manager 5.3 software (Cochrane Community). The heterogeneity was assessed with the Mantel-Haenszel method, and the I 2 values were also calculated. The I 2 was considered low (<25%), moderate (25-50%), or high (>50%). A fixed-effects model was used when I 2 <50%, whereas a random-effects model was used when I 2 >50%. The publication bias was also assessed using funnel plots.

Characteristics of included studies
A total of 1,504 articles were retrieved from three databases during the search, and eight of these articles, including retrospective studies, prospective studies, and case series, were included in this study (Fig 1). The characteristics of the eight studies are shown in Table 1

Prevalence of severe P. ovale infections
Among the eight included studies, which included a total of 1,365 ovale malaria cases, the pooled prevalence of severe P. ovale was 0.03 (95% CI = 0.03-0.05%, I 2 = 54.4%). The highest proportions of severe P. ovale were found in the study by de Laval et al.  (Fig 2). According to the results of the meta-regression of six included studies, the age of the participants did not confound to the pooled prevalence of severe P. ovale infections (P-value = 0.3, 95% CI = -0.002-0.005). The complications most commonly found in patients with P. ovale infections were jaundice (1.1%), severe anemia (0.88%), and pulmonary impairments (0.59%) ( Table 2).

PLOS ONE
Severe P. ovale infection

Mortality rates of P. ovale and Plasmodium spp. infections
There was only one study that reported death cases resulting from P. ovale infections [19]. The mortality rate of severe P. ovale infections in their study was 0.22% (2/903). The mortality rate of severe P. ovale infections among the eight included studies was 0.15% (2/1,365

Quality of included studies
Four studies were rated as having 9 stars (good quality), three studies were rated as having 7 stars (adequate quality), and one study was rated as having 6 stars (adequate quality) according to the NOS concerning the selection process for the cases and controls included ( Table 3).

PLOS ONE
Severe P. ovale infection

Publication bias
A funnel plot analysis of the included studies could not be performed to assess the publication bias among the included studies, as a minimum of 10 studies were required for the analysis [39].

Discussion
The present systematic review and meta-analysis aimed to explore the prevalence of severe P. ovale infections and to summarize the mortality caused by P. ovale. The results demonstrated that 3% of patients infected with P. ovale developed severe complications according to the WHO 2014 guidelines [1]. Therefore, the term "benign malaria" might not apply to patients tested positive with P. ovale, especially travelers after returning from countries in which P. ovale is endemic. In addition, international and national healthcare providers should be recommending the use of chemoprophylaxis for travelers, as the present study revealed that P. ovale, long-recognized as benign malaria, can cause severe and fatal clinical manifestations. In the present study, a meta-regression with age as a covariate was performed, and the pooled prevalence of severe P. ovale infections was investigated. The results showed that the age of the participants in six of the included studies did not confound the pooled prevalence of severe P. ovale infections. A previous study demonstrated that the highest incidence of P. ovale was observed among children 0-7 years old, who had an 8-fold higher risk of P. ovale infections than did adults; nevertheless, clinical attacks were observed in all age groups [40]. The present study also demonstrated a significant difference between the proportion of patients with severe P. ovale infections and severe P. falciparum infections. A lower proportion of patients with severe P. ovale infections than of patients with P. falciparum infections had severe cases. This finding indicated that although P. ovale can cause severe malaria, the chance of patients developing severe complications is lower with P. ovale than with P. falciparum, which is the Plasmodium species widely known to be the leading cause of severe malaria in humans. This study also demonstrated there are no significant differences in the proportions of patients with severe P. ovale and with P. vivax/P. malariae. This finding demonstrated that there are relatively few severe P. ovale cases, including P. vivax and P. ovale cases. In contrast to P. vivax, which is endemic in Asia, Central America, and South America, P. ovale is only endemic in some African countries [3]. However, the study conducted by Langford et al. in 2015 [34] indicated that there were 120 cases of P. ovale infection in Indonesia, whereas other included studies reported that travelers who returned to their home countries had imported P. ovale malaria cases [19,32,[35][36][37][38].
Although microscopy is still considered to be the gold standard for the identification of Plasmodium species, the morphological characteristics of P. ovale are similar to those of P. vivax and are readily missed by less experienced microscopists [41,42] [43]. In addition, missed diagnoses of P. ovale infections leading to relapses with a low parasitemia level is a problem related to this Plasmodium species [44]. Previous studies have indicated that the sensitivity of routine microscopy in detecting P. ovale in imported cases is very low and related to low parasitemia levels [32,45]. Moreover, the rapid diagnosis tests (RDTs) that have been developed and are available lack the sensitivity required to detect low amounts of the circulating antigen of P. ovale [46,47]. Moreover, a previous study indicated that routine microscopic examinations with thick and thin blood smears should be repeated three times in cases of suspected imported P. ovale malaria [48]. Ideally, molecular detection using polymerase chain reaction (PCR) to detect and confirm P. ovale infections in cases of low parasitemia or negative blood film is needed, although it is not routinely available [32,43]. Therefore, the limitations of routine microscopic and RDT tests might affect the management/treatment of patients and lead to an increase in the morbidity associated with P. ovale infections [49]. The mortality caused by severe P. ovale infections among travelers who returned to the USA was previously reported by Hwang et al. in 2014 [19]. The authors recommended the use of suppressive prophylaxis and antirelapse treatment with primaquine drugs for patients who returned from ovale-endemic countries [19]. Nevertheless, a previous case report showed that severe complications led to fatality in patients who had received anti-malarial prophylaxis treatment during their trip to Nigeria [20]. This finding may have been caused by the impropriate usage of prophylactic drugs or the survival of hypnozoites during anti-malaria chemoprophylaxis against the P. ovale infection [50].
The mechanism by which P. ovale infections lead to death is still unknown, but acute renal failure and acute respiratory distress syndrome might act as secondary contributing factors to death [51]. People who previously acquired P. falciparum malaria and those staying in malaria-endemic regions are known to show less severe symptoms during subsequent malaria infections than are people who are from non-malaria endemic regions. A previous study investigated two cases of P. ovale in patients, and it was suggested that one patient with a history of malaria was protected against developing severe complications, while the other patient, who was experiencing malaria infection for the first time, suffered severe complications [51]. Therefore, severe P. ovale infections can occur in nonimmune individuals in particular.
This systematic review indicated that jaundice, severe anemia, and pulmonary impairment are the severe complications most commonly found in patients infected with P. ovale. For jaundice, a previous study indicated that mild hyperbilirubinemia can be found in approximately 50% of patients infected with P. ovale [52]. Jaundice concomitant with P. ovale infection might be due to liver dysfunction to conjugate bilirubin in cases of severe red blood cell destruction during malaria infection. Severe anemia concomitant with P. ovale infection might occur in patients with underlying diseases such as hemoglobinopathies, causing the severe destruction of abnormal red blood cells. The mechanism of pulmonary impairment in patients with P. ovale malaria is not clear. In patients with P. falciparum infection, cytoadherence, mechanical obstruction, and inflammation of the microvascular endothelium at the pulmonary vasculature due to infected red blood cells that cytoadhere to the microvascular endothelium causes the mechanical obstruction of pulmonary vasculature; thus, alveolar capillary permeability increases, and intravascular fluid is spread into the lungs, leading to pulmonary failure [53].

Summary of the evidence
The present systematic review and meta-analysis presented cases of severe manifestations and demonstrated the severity and mortality of severe P. ovale infection to increase the awareness of physicians regarding the prognosis of this disease. It is of the utmost importance to test individuals for malaria after they return from malaria-endemic areas, as severe P. ovale complications could occur, especially in nonimmune travelers. Last, the knowledge on appropriate management strategies and the necessary interventions concerning malaria diagnosis affects the severity of malaria manifestations and patient outcomes. Clinicians' and technicians' familiarity with the endemicity of malaria in an area or a region therefore plays an important role in the initial stages of laboratory assessments and subsequent diagnoses.

Limitations
A limitation of this systematic review and meta-analysis was that there were a limited number of available publications on severe complications and death related to P. ovale infection. Another limitation was that results from case series and case-control studies were used for prevalence estimations. However, previous studies XXX have suggested and used case reports/ series in meta-analyses to facilitate the decision making process when the evidence was limited or a relatively rare or neglected disease was assessed [54][55][56][57][58][59]. In addition, the most recently published meta-analysis of 44 case series reported the 25-year pooled survival of hip replacements [60]. Considering these limitations, the meta-analysis results and conclusions on the prevalence of severity and mortality associated with P. ovale infection should be considered with caution in combination with the results of newly published reports.

Conclusion
This systematic review demonstrated that although P. ovale infections have long been considered cases of benign malaria, severe complications in patients have been reported on rare occasions. The possible reasons for the small number of reports on the severity of the malaria infections linked or caused directly by P. ovale should be studied further by the scientific community. Clinicians and technicians need to recognize that patients who return from malariaendemic areas may develop severe P. ovale infections. Additional studies need to consider the potential underreported presence of P. ovale malaria infections in the clinical setting so that appropriate management strategies and interventions can be administered to infected individuals and the knowledge on malaria epidemiology can be expanded further.
Supporting information S1 Checklist. PRISMA checklist. PRISMA statement for reporting systematic reviews and meta-analyses. (DOC) S1