Figures
Abstract
Background
Sub-microscopic (SM) Plasmodium infections represent transmission reservoirs that could jeopardise malaria elimination goals. A better understanding of the epidemiology of these infections and factors contributing to their occurrence will inform effective elimination strategies. While the epidemiology of SM P. falciparum infections has been documented, that of SM P. vivax infections has not been summarised. The objective of this study is to address this deficiency.
Methodology/Principal Findings
A systematic search of PubMed was conducted, and results of both light microscopy (LM) and polymerase chain reaction (PCR)-based diagnostic tests for P. vivax from 44 cross-sectional surveys or screening studies of clinical malaria suspects were analysed. Analysis revealed that SM P. vivax is prevalent across different geographic areas with varying transmission intensities. On average, the prevalence of SM P. vivax in cross-sectional surveys was 10.9%, constituting 67.0% of all P. vivax infections detected by PCR. The relative proportion of SM P. vivax is significantly higher than that of the sympatric P. falciparum in these settings. A positive relationship exists between PCR and LM P. vivax prevalence, while there is a negative relationship between the proportion of SM P. vivax and the LM prevalence for P. vivax. Amongst clinical malaria suspects, however, SM P. vivax was not identified.
Conclusions/Significance
SM P. vivax is prevalent across different geographic areas, particularly areas with relatively low transmission intensity. Diagnostic tools with sensitivity greater than that of LM are required for detecting these infection reservoirs. In contrast, SM P. vivax is not prevalent in clinical malaria suspects, supporting the recommended use of quality LM and rapid diagnostic tests in clinical case management. These findings enable malaria control and elimination programs to estimate the prevalence and proportion of SM P. vivax infections in their settings, and develop appropriate elimination strategies to tackle SM P. vivax to interrupt transmission.
Author Summary
Light microscopy (LM) has been the mainstay of malaria diagnosis for case management and surveillance. The introduction of molecular based diagnostics such as polymerase chain reaction (PCR) in the 1990s resulted in increased reporting of Plasmodium infections in communities compared to LM, indicating that sub-microscopic (SM) Plasmodium infections are endemic in a variety of different settings. The prevalence and transmission potential of SM P. falciparum was reviewed; however, information on the prevalence and distribution of SM P. vivax in different settings is not available. In this article, the authors analysed LM and PCR results described in 38 publications (44 studies), and revealed that SM P. vivax is highly prevalent across different geographic areas with varying transmission intensities. On average, these infections made up 67% of P. vivax infections detected by PCR in cross-sectional surveys, and were relatively more prevalent in areas where malaria was under control and moving towards elimination. By contrast, SM P. vivax is not prevalent in clinical malaria suspects. These findings highlight that for detection of clinical malaria LM and rapid diagnostic tests (RDTs) are adequate, however, to detect very low density infections that are largely asymptomatic but still contribute to transmission, more sensitive diagnostic tools are needed.
Citation: Cheng Q, Cunningham J, Gatton ML (2015) Systematic Review of Sub-microscopic P. vivax Infections: Prevalence and Determining Factors. PLoS Negl Trop Dis 9(1): e3413. https://doi.org/10.1371/journal.pntd.0003413
Editor: Mauricio Martins Rodrigues, Federal University of Sao Paulo, Brazil
Received: September 2, 2014; Accepted: November 12, 2014; Published: January 8, 2015
Copyright: © 2015 Cheng et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. Data are available from published articles as cited in the manuscript.
Funding: QC is funded by Department of Defence, Australia; JC by the World Health Organization through a grant from UNITAID; and MLG by the Queensland University of Technology. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The opinions expressed herein are those of the authors and do not necessarily reflect those of the Australian Defence Force Joint Health Command or the World Health Organization.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The global malaria incidence and death rates have decreased in recent years [1] due to increasing funding and political commitment, as well as implementation of artemisinin combination therapy, better access to diagnostics and vector control interventions such as insecticide treated bed nets and indoor residual spray. As a result, many countries/regions are planning or have already committed to eliminating malaria. In these areas the malaria surveillance programs that generate information on malaria cases, burden and transmission trends need to be strengthened and extended to include case and foci investigations [2]. These focused investigations play a pivotal role in informing malaria elimination action plans and directing resources.
The effectiveness of malaria surveillance depends on the performance of surveillance tools. Light microscopy (LM) has been the main surveillance tool over the past decades. LM provides important epidemiological information such as malaria incidence rates, burden and relative species composition. However, since the introduction of molecular based diagnostics, e.g. polymerase chain reaction (PCR) in the 1990s, there has been increased reporting of malaria infections in communities which are detected by PCR, but not by LM. These reports demonstrate LM's limitation in detecting infections with low parasite densities; levels well below the threshold for symptomatic malaria, but of sufficient density to enable transmission of parasites to mosquitoes [3], [4]. Therefore, it is important to understand the epidemiology of these sub-microscopic (SM) malaria infections in different settings and their role in maintaining malaria transmission, particularly in the context of elimination strategies.
SM P. falciparum infections are well documented. A systematic review and meta-analysis of P. falciparum LM and PCR prevalence data revealed that on average, the prevalence of LM was only 50.8% of that measured by PCR [5]. This suggests that half of all detected P. falciparum infections were SM. The meta-analysis revealed that the SM P. falciparum is more common in adults, in areas with low transmission intensities and in chronic infections [6]. Based on findings of earlier experimental studies, Okell and colleagues estimated that SM P. falciparum contributes between 20% and 50% of human to mosquito transmission in areas with low and very low transmission intensity [6]. This has great implications for malaria control and elimination programs because SM P. falciparum infections cannot be readily detected by diagnostic tools commonly used for case management or field surveillance (such as LM and rapid diagnostic tests, RDTs) and these undetected large reservoirs of SM P. falciparum can maintain low level of transmission and seed outbreaks [7].
In contrast to P. falciparum, the prevalence and distribution of SM P. vivax has not been systematically analysed despite increasing reports of asymptomatic and SM P. vivax. This is a major gap in moving forward with malaria elimination in certain regions since it is likely that the epidemiology of SM P. vivax is different to that of P. falciparum due to several unique biological features of P. vivax. For example, parasite invasion of Duffy positive reticulocytes is thought to contribute to the observed overall lower parasite densities in P. vivax infections compared to P. falciparum infections, thus heightening the theoretical likelihood of having SM infections. Other features such as relapses may affect the speed at which host immunity develops, which in turn may affect parasite density. Mature gametocytes are present earlier in P. vivax infections meaning they can infect mosquitoes at an early stage of infection, with SM P. vivax gametocytes shown to be able to successfully infect mosquitoes [8], [9]. Dormant P. vivax parasite stages in the liver (hypnozoites) that can activate at variable periodicity depending on geographical region [10], means that untreated SM infections will continue to relapse in the future with the potential to continue transmission during each relapse. Therefore, SM P. vivax poses a major challenge to malaria control and elimination programs in areas where P. vivax is endemic.
In this study we reviewed published literature and analysed the prevalence and relative proportion of SM P. vivax infections across different transmission settings. We also investigated factors associated with the occurrence of SM P. vivax infections. These findings can support the reorientation of malaria control programmes towards elimination of P. vivax.
Methods
Literature search
A literature search was conducted in PubMed using the search terms “vivax, PCR, survey” in Jan 2014. These initial publications (excluding 3 non-English papers) were then carefully reviewed and selected according to the following inclusion criteria: 1) Data were collected either from cross-sectional surveys of a population or a representative subset of population at one point in time or from screening studies of clinical malaria suspects, 2) Data include separate microscopy and PCR-based results for P. vivax in the same setting (RDT results were not considered), and 3) At least one P. vivax infection was detected by either PCR or LM.
Data analysis
When results of multiple surveys were reported in a single paper, either in different locations or in different season, data from these surveys were combined if the authors did not report a significant difference in prevalence between different locations or seasons. However, if prevalence was reported to be significantly different between locations or seasons by the authors each survey was included separately, unless 1) the number of samples in each location/season was <50 and 2) there were no data provided for each individual location/season.
Terminology
- Microscopy method refers to detection of Plasmodium spp. using light microscopy (LM) examination of thick and thin blood smears following the WHO recommended protocol.
- PCR method refers to detection of Plasmodium spp. by amplification of any parasite gene using any PCR-based technology, including conventional single round, multiplex, nested, semi-nested PCR, real time quantitative PCR and ligase detection reaction-fluorescent microsphere assay (LDR-FMA).
- Parasite prevalence determined by LM and PCR. This was calculated as: total number of positive samples/total number of samples examined ×100%.
- Prevalence of sub-microscopic (SM) infections. The SM prevalence was calculated as: PCR prevalence – LM prevalence. In surveys where LM prevalence was higher than PCR prevalence, the prevalence of SM is considered as 0.
- Relative proportion of sub-microscopic (SM) infections. This was calculated as: SM prevalence/PCR prevalence ×100%.
Statistical tests.
Paired comparisons of LM and PCR, and P. falciparum and P. vivax SM prevalence or SM proportion for all sites were tested using Wilcoxon matched pairs signed rank test (GraphPad Prism). Comparisons of LM, PCR and SM prevalence or relative SM proportions between different age groups and laboratory methods were performed using the Mann Whitney test.
Regression analysis.
The relationship between LM and PCR prevalence was assessed using linear regression of the log10 transformed values. Log transformed values were used to resolve heteroscedasticity. The relationship between LM prevalence and proportion of SM infections was analysed using a generalized linear model with a gamma distribution and log link function.
Results
Search outcome and data grouping
The PubMed search produced a list of 139 publications, of which 38 met the inclusion criteria. Twenty five publications [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35] reported findings of 31 cross-sectional surveys of different populations (including 29 cross-sectional surveys, one reactive case investigation and one cohort study) that were conducted by household or village based or random sampling. The remaining 13 publications [18], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47] reported findings of fever/or clinically suspected malaria patient screening. LM and PCR results reported in these two groups of publications were analysed separately. The location, year and references of these studies are summarised in Table 1.
Prevalence of P. vivax and P. falciparum infections in communities
The 31 cross-sectional community surveys were conducted in 12 countries (Table 1). Twenty eight of these surveys were conducted between 1996 and 2010. Survey year was not described for the remaining three surveys (Table 1). The number of samples tested by PCR in each survey ranged from 98 to 3316 (median = 337, interquartile range: 252 to 1269). The LM prevalence of P. vivax ranged from 0.0% to 44.3% in geographically different settings, while the corresponding PCR prevalence for P. vivax in these same settings ranged from 0.2% to 59.5%. Overall across all sites, the PCR prevalence of P. vivax was significantly higher than that of LM (Wilcoxon matched-pairs signed rank test, P<0.0001, Fig. 1), i.e. PCR detected a significantly higher number of P. vivax infections than LM.
The total height of each bar (blue + red) represents the PCR prevalence. Countries where data were collected and their corresponding references (detailed in Table 1) are shown on the x-axis.
The prevalence of sympatric P. falciparum, including P. falciparum in mixed species infections, was also analysed for 30 of the 31 surveys as a comparison. Prevalence of P. falciparum determined by LM ranged from 0.0 to 40.4% while PCR prevalence of P. falciparum PCR ranged from 0.0 to 81.5%. Similar to P. vivax, the overall prevalence of P. falciparum determined by PCR was significantly higher than that of LM (Wilcoxon matched-pairs signed rank test, P<0.0001, Fig. 2).
Total height of each bar (blue + red) represents the PCR prevalence. Countries where data were collected and their corresponding references (detailed in Table 1) are shown on the x-axis.
Prevalence and relative proportion of SM P. vivax and P. falciparum infections
The prevalence of SM P. vivax in the 31 community surveys analysed ranged from 0.2 to 48.6% (Fig. 1). The relative proportion of P. vivax not detected by LM (i.e. SM infections) ranged from 1.5% to 100.0%, with a mean of 69.5% (Fig. 3). The prevalence of SM P. falciparum ranged from 0.0% to 61.3%, constituting, on average, 55.7% of P. falciparum infections (range 0.0% to 100.0%, Fig. 4).
Countries where data were collected and their corresponding references (detailed in Table 1) are shown on the x-axis.
Countries where data were collected and their corresponding references (detailed in Table 1) are shown on the x-axis.
The prevalence of SM P. vivax in these communities was not significantly different to that of SM P. falciparum (Wilcoxon matched-pairs signed rank test, P = 0.78, Fig. 5A). However, the average relative proportion of SM was significantly higher in P. vivax infections compared to P. falciparum in the same study (Wilcoxon matched-pairs signed rank test, Median difference = 5.2%, P = 0.045, Fig. 5B).
B) Comparison of relative proportions of SM (sub-microscopy) P. vivax and P. falciparum (mean with 95% CI) in 31 cross-sectional surveys.
Relationships between LM and PCR determined P. vivax prevalence
A positive relationship between LM and PCR determined P. vivax prevalence was observed (Fig. 6). LM P. vivax prevalence was a significant factor for predicting PCR P. vivax prevalence in community surveys (P<0.001, R2 = 0.675). The fitted regression equation is: log (PCR P. vivax prevalence) = 0.596× log (LM P. vivax prevalence) – 0.003. Thus, PCR detectable P. vivax prevalence can be estimated based on LM P. vivax prevalence for community surveys using quality LM and PCR with similar sensitivities to those reported in the studies described here.
Relationship between LM and relative proportion of SM P. vivax
A negative relationship between LM P. vivax prevalence and proportion of SM P. vivax infections was identified (Fig. 7), with LM P. vivax prevalence identified as a significant factor for predicting the proportion of samples that are PCR positive/LM negative (SM) in cross-sectional surveys where the LM P. vivax prevalence is less than 45% (P<0.001, Deviance = 7.32, df = 29). The regression equation describing this relationship is: proportion SM P. vivax = exp (−0.162–4.163× LM P. vivax prevalence). The negative relationship observed remains when a furthest outlier was removed from the analysis.
Pathogenesis of SM P. vivax
The pathogenesis of SM could be indicated by the percentage of infected individuals having malaria related symptoms at the time of the survey. Malaria symptoms include acute symptoms normally represented by fever, and chronic symptoms represented by anaemia. Data on the proportion of symptomatic P. vivax mono infections in PCR positive/LM negative subjects could only be extracted from six studies [16], [22], [25], [26], [27], [35], all of which were conducted in areas with relatively low transmission (LM P. vivax prevalence ranged 0.01% to 4.79%). The proportion of individuals with symptoms ranged from 0.0% to 11.4%, averaging 2.8%. This means that between 88.6 and 100.0% (average 97.1%) of individuals with SM P. vivax were asymptomatic. To ascertain whether these asymptomatic SM carriers will become symptomatic some time later, one study followed 25 LM negative/PCR positive subjects in South America for two months and found they remained asymptomatic over the 2 month duration [26].
The relationship between SM P. vivax and anaemia was investigated by Ladeia-Andrade et al. [23] in Brazil using two-level logistic regression models. After excluding LM positive subjects, and controlling for covariates such as age and sex, they concluded that the presence of SM P. vivax was a significant predictor of anaemia (OR = 1.92; 95% CI:1.14–3.23; P = 0.015).
Factors contributing to the occurrence of SM infections
- Anti-parasite antibodies. Antibodies against P. vivax are a marker of host immunity developed during previous infections. Four cross-sectional surveys [22], [25], [26], [28] included a serological investigation of anti-P. vivax antibodies using ELISA or IFA in parallel with LM and PCR. However, only two studies compared serological and PCR results. Versiani et al [26] observed that PCR positive/LM negative subjects were 2.45 fold more likely to have anti-PvMSP1 antibodies than PCR negative/LM negative subjects in Amazonas State of Brazil, and that the PCR positive/LM negative group had significantly higher titres of anti-PvMSP1 antibodies than the PCR negative/LM negative group [26]. Increased presence of anti-PvMSP1 antibodies was also noted for PCR positive/LM positive subjects (compared to subjects with no evidence of P. vivax infection), however there were too few cases positive by both PCR and LM (n = 15) to achieve statistical significance, or compare between SM and patent infections. A second study conducted in Bellavista, Peru observed that PCR positivity was significantly associated with the presence of anti-PvMSP1 antibodies [28], however no comparison was made between SM and patent infections.
- Age. In many malaria endemic settings, the adolescent population has higher proportions of asymptomatic and low density Plasmodium infections than young children due to the development of clinical immunity after repeated exposure to P. vivax parasites. Therefore, age is a surrogate marker for acquired clinical immunity. Fifteen of the 25 publications (21 of the 31 surveys) described prevalence of P. vivax infection in different age groups, but only 12 publications (16 surveys) analysed the relationships between P. vivax LM or PCR prevalence and age. Seven publications (eight surveys) found no significant association between the relative proportion of SM P. vivax infection and host age [14], [15], [17], [22], [27], [28], [30], while five publications (eight surveys) reported that the relative proportion of SM P. vivax increased with age [12], [23], [31], [32], [34]. Neither the LM nor the PCR P. vivax prevalence was significantly different between the studies that did and did not find an association with age (Mann Whitney test, P>0.05).
- Microscopy QA and DNA extraction method for PCR. The prevalence and relative proportion of SM P. vivax could vary because of differences in quality of microscopy and PCR. Only one [35] of the 25 publications (1 of the 31 surveys) stated the competency levels of microscopists according to the WHO accredited malaria microscopy competency assessment. Eleven publications (15 surveys) described the microscopists as experts, experienced, highly skilled or well-trained but did not provide information on qualifications. Fifteen of the 31 surveys reported some form of quality assurance (QA) of the microscopy including two independent microscopists reading and/or an expert referee to confirm positives, discordant slides and random selection of negatives, as well as PCR positive results. However, the presence or absence of a description for performing microscopy QA did not affect the average prevalence of SM (Mann Whitney test, P = 0.2447) or the average proportion of SM P. vivax (Mann Whitney test, P = 0.1626).
While quality of microscopy largely depends on the competency of microscopists, quality of PCR can be particularly influenced by the type and volume of blood used to extract DNA. All surveys described the type of blood and DNA extraction method used for PCR analysis. Fourteen surveys used blood from filter paper while 17 surveys used whole blood (primarily>150 µL) for DNA extraction. The average LM prevalence of P. vivax was not significantly different between these two methods of DNA extraction (Mann Whitney test, P = 0.161). However, the average prevalence of SM P. vivax in community surveys determined by PCR using whole blood was significantly higher than that using blood from filter papers (Mann Whitney test, P = 0.004). - Fever and drug use. Antimalarial drug use could also affect parasite density at the time of survey and thereby increase the prevalence and relative proportion of SM P. vivax. Only one survey reported that recent antimalarial treatment (<4 weeks prior to survey) was associated with a significant increase in risk of having SM P. vivax infection [33].
Prevalence of SM P. vivax infections in surveys among clinical malaria suspects and their proportions in different settings
In 13 surveys of clinical malaria suspects conducted in different transmission settings (1997–2010), the prevalence of P. vivax infections determined by LM varied widely from 0.7% to 86.0% of the patients. Although PCR detected more P. vivax infections in some studies, overall it did not have a significantly higher prevalence than LM in fever patients (Wilcoxon matched-pairs signed rank test, P = 0.278, Fig. 8). In these same surveys, the LM P. falciparum prevalence rates ranged from 1.0% to 54.2% while PCR prevalence rates ranged from 1.0% to 46.9%. Similar to P. vivax, PCR did not detect a significantly higher number of P. falciparum infections than LM in these patients (Wilcoxon matched-pairs signed rank test, P = 0.123, Fig. 9). This suggests that amongst symptomatic patient populations SM P. vivax and P. falciparum are not prevalent.
Total height of the bar (blue + red) represents the PCR prevalence.
Total height of bar (blue + red) represents PCR prevalence.
Discussion
Sub-microscopic malaria infections have been increasingly reported in malaria endemic areas, especially in places where malaria transmission intensities are relatively low and where malaria elimination is being targeted. In this paper, published articles from varying transmission intensities were reviewed and data extracted for secondary analysis with the purpose of providing an overview of SM P. vivax malaria in different settings including its prevalence and relative proportion of all detectable P. vivax infections, and relationship with LM prevalence, as well as consideration of the factors contributing to SM P. vivax.
The 31 cross-sectional surveys reviewed were conducted in areas where LM P. vivax prevalence ranged from 0 to 44%. Since entomological inoculation rate (EIR) was not reported in these surveys, we used LM P. vivax prevalence as an indicator of transmission intensity in each setting. One of the advantages of using LM prevalence instead of PCR prevalence is that LM prevalence has been used as a measure of transmission intensity in endemic countries for the past century. Our analysis showed that the PCR P. vivax prevalence was significantly higher than that LM P. vivax prevalence in these surveys, demonstrating the presence of a sub-population carrying SM P. vivax across all transmission settings. The prevalence of SM P. vivax infections ranged from 0.2% to 48.6%. SM P. vivax constitutes between 1.5% and 100% of all P. vivax infections detected by PCR (average 69.5%) in these settings, and this proportion is significantly higher than that of the sympatric P. falciparum infections. This suggests that without PCR or other comparably sensitive methods, an average of 69% of P. vivax infections would be undetected hence antimalarial treatment will usually not be sought. This results in a potentially large group of people carrying parasites that have the potential to transmit P. vivax [8], [9] and maintain the transmission cycle. Furthermore, the majority of these people are asymptomatic and may remain so for months [26].
Similar to the epidemiology of SM P. falciparum reported by Okell et al.[6], a negative relationship was identified between the relative proportion of SM P. vivax and LM P. vivax prevalence in 31 cross-sectional surveys. This suggests that a relatively larger proportion of P. vivax infections are SM in areas with low transmission intensity compared to areas with high transmission intensity, although SM P. vivax is prevalent in both type of settings. Hence, SM P. vivax poses more challenges to malaria control programs in areas where transmission intensities are already low and are progressing toward elimination.
Many factors can contribute to the occurrence and level of SM P. vivax. These include technical, parasite and host factors. From a technical point of view, the quality of microscopy is an important factor that can contribute to the level of LM P. vivax, and thus SM P. vivax. The prevalence and relative proportion of SM P. vivax would increase if the microscopist is relatively poor at detecting parasites, or decrease if there are false positive results (eg cell debris wrongly classified as a parasite). The correct speciation of parasites is also important in areas with both P. falciparum and P. vivax. This is a special concern since quality of field microscopy has been shown to be highly variable [48], [49], [50], [51], [52], [53], [54]. Interestingly, the description of microscopy QA was not found to be associated with the prevalence or relative proportion of SM P. vivax in the 31 cross-sectional surveys reviewed. This indicates that the quality of microscopy of these surveys was comparable between studies, irrespective of whether QA was reported or not.
Quality of PCR is also an important determinant for the level of SM P. vivax. It might be expected that a sensitive PCR would detect more SM P. vivax than a less sensitive PCR in a field survey. The PCR sensitivity is largely determined by the quality and amount of parasite DNA and by copy number of the target gene. Of the 31 cross-sectional surveys, 29 (93.5%) used a PCR-based assay targeting the parasite 18sRNA gene. Therefore, the sensitivity of these PCR methods would most likely be affected by the number of parasites added in each assay which is determined by the volume of blood used for DNA extraction and the concentration of parasite DNA. Seventeen of the 31 cross-sectional surveys used >150 µL of whole blood while 14 surveys used dried blood spot on filter paper which usually contain 5-20 µL of blood. As expected, the surveys using whole blood produced a significantly higher average prevalence of SM P. vivax compared to the surveys using filter paper blood. Although the final volume of DNA elusion could not be assessed, this difference is likely due to the higher number of parasites present in the larger volume of whole blood, resulting in more parasite DNA being added into each assay. This finding suggests that while filter paper helps to preserve blood in field conditions and assists with transportation of samples, the method is less sensitive than whole blood in detecting SM P. vivax. In order to maximize PCR sensitivity in detecting SM parasite infections, >150 µL of whole blood should be used when possible.
Parasite factors influencing SM P. vivax may include the growth characteristics and virulence of parasite strains, as well as genetic diversity, susceptibility to antimalarial drugs and relatedness of parasite strains. While the former, which is difficult to study in the field, may directly affect the parasite density, the latter could indirectly affect the parasite density through effective host immunity. For example, homogeneity of parasite population, i.e. lack of genetic diversity, is often reported in areas with low transmission intensity, such as Peru and South American countries [55], [56], [57]. This limited parasite genetic diversity could speed up the development of acquired immunity in a host population, which in turn could reduce parasite density in the host resulting in a higher proportion of SM infections. However, parasite diversity was not reported as part of any of the surveys reviewed. Recently, Gray et al [58] reported genetic diversity and parasite relatedness in Temotu, Solomon Islands as a follow up of the baseline survey [35]. In the baseline survey 75% of LM positive P. vivax infections had parasite densities below 100/µL with a further 44% of P. vivax infections being SM P. vivax; 94% of all P. vivax infections were asymptomatic at the baseline survey. However, genetic diversity of P. vivax population was very high and parasite haplotypes were not highly related [58]. This suggests that genetic diversity of P. vivax was not a major contributor to the high prevalence of SM P. vivax in this setting.
Host factors including non-specific immune responses, immunity status and human genetics could greatly impact parasite density, and hence the prevalence and proportion of SM P. vivax. Firstly, the non-specific immune response commonly associated with febrile illness has been hypothesized to directly reduce parasite density [59], [60] and any resultant antimalarial treatment can rapidly eliminate parasites. The effect of febrile illness and antimalarial treatment on SM P. vivax was investigated in one of the 32 cross-sectional surveys reviewed [33]. Lin et al. reported that fever episodes in the two weeks prior to sample collection, and antimalarial treatment 4 weeks prior, were associated with a significant reduction in risk of being detected by LM or PCR-based method [33]. Therefore, antimalarial treatment reduces both LM and SM P. vivax infections.
Host immune status could also be a major determinant of SM P. vivax. It has been reported that SM P. falciparum is more common in adults compared to children [6]. However, this pattern was not exactly repeated in P. vivax. Of the 16 surveys that analysed the relationship of SM P. vivax with age, eight reported the relative proportion of SM P. vivax was not associated with age, while the other eight reported that the relative proportion of SM P. vivax increased with age. This difference may be related to the transmission intensities in study regions, however a comparison of the average LM P. vivax prevalence between these two groups did not show a significant difference. One other possibility may be difference in geographical or host population; the group of surveys reporting no association with age were mostly conducted in Southeast Asia and South America, while those reporting an association included four surveys conducted in PNG. The presence and quantity of anti-PvMSP1 antibodies have also been associated with SM P. vivax in two surveys conducted in South America [26],[28]. This suggests that serology may help detect SM P. vivax carriers.
Associations between other host factors such as polymorphisms in Duffy antigen, haemoglobin or G6PD and SM P. vivax infections were not investigated in any of the 31 cross-sectional surveys reviewed. Further studies are required to ascertain the contribution of host genetics to SM P. vivax infections.
Based on findings of six surveys, between 89% and 100% (average 97.5%) of SM P. vivax infected subjects were asymptomatic at the time of survey and it was not well documented whether these people progressed to develop LM detectable parasitemia at a later time. One study followed 25 LM-/PCR+ subjects and found them all asymptomatic after a 2 month follow up [26]. If these individuals remain asymptomatic then they would not be identified by either active or passive case detection based on LM or RDTs with a similar sensitivity to LM.
Any individual infected with blood stage P. vivax, regardless of the level of parasitemia, is likely carrying dormant hyponozoites in the liver and is thus likely to relapse weeks to months after the primary infection. Future relapses in these individuals can only be limited by treatment that includes antimalarials (specifically, primaquine) targeting the P. vivax hypnozoites. Therefore, to prevent relapse WHO recommends that in low transmission areas patients with P. vivax infection, who are not G6PD deficient, receive treatment against both blood (such as chloroquine) and hypnozoite (primaquine, 0.25 or 0.5mg/kg/day once a day for 14 days) stages [61]. However, SM P. vivax infections are not detectable using LM or RDTs, as such these infected individuals would not receive any anti-malarial therapy for P. vivax blood or liver stages and thus continue to transmit P. vivax. The prevalence of this undetectable and untreated SM P. vivax population creates a major challenge to malaria control and elimination programs. Strategies for detecting and treating infected cases (patent and symptomatic infections) alone will unlikely interrupt transmission because SM P. vivax infections and their future relapses will continue to feed into transmission. Comparing to SM P. falciparum, SM P. vivax infections represent a more important transmission reservoir due to multiple relapses occurring over a long period of time. To accelerate malaria elimination i.e. the complete interruption of malaria transmission and total removal of the disease burden of malaria such as anaemia, some form of mass screening using PCR-based or comparably sensitive methods and radical cure approach would be required to identify and treat all SM P. vivax infected subjects. This strategy would be expected to have more impact on P. vivax transmission than on P. falciparum transmission because appropriate management of SM P. vivax will not only stop transmission from current SM infections, but also prevent future relapses and transmissions resulting from these relapses. However, because the conventional PCR based assays are expensive to implement and difficult to perform under field conditions, mass screening will not be practical before more sensitive and specific field deployable diagnostic tests become available. Cost-effectiveness studies will be needed to properly evaluate the options available for P. vivax elimination strategies for different transmission settings. In the interim, the PCR prevalence of P. vivax could be estimated based on the positive relationship between PCR P. vivax prevalence and LM P. vivax prevalence identified in this study. If the prevalence of PCR positive P. vivax is much higher than that of LM, mass drug administration may be an option for elimination but it also has its operational challenges, risks and potential limitations [62].
In contrast to cross-sectional surveys, SM P. vivax was much less prevalent in clinical malaria suspects. Overall, the PCR prevalence for P. vivax or P. falciparum was not significantly higher than that of LM in 13 surveys of febrile patients. This could be due to relatively high parasite density in symptomatic patients. This re-emphasizes that quality LM and RDTs are adequate tools for case management of both P. vivax and P. falciparum patients.
In summary, SM P. vivax is prevalent across different geographic areas with varying transmission intensities constituting, on average, 69.5% of all P. vivax infections. The relative proportion of SM P. vivax is significantly higher than that of the sympatric P. falciparum in these settings and is higher in areas with relatively low transmission intensity. These SM P. vivax infections not only have negative health impact on the infected individual, but will also contribute to P. vivax transmission both from the current infection and subsequent relapses, and thus present a major challenge for malaria elimination programs. This review seeks to provide malaria control and elimination programs with estimates of the prevalence and proportion of SM P. vivax infections in their settings, and to highlight the importance of developing diagnostic tools for detecting SM P. vivax infections in order to support elimination strategies. Strategies for tackling both patent and SM P. vivax are critical for eliminating P. vivax.
Acknowledgments
The authors would like to thank Dr. James McCarthy for his assistance in obtaining a list of references on P. vivax transmission and Dr. Andrea Bosman for critical review of the manuscript.
Author Contributions
Conceived and designed the experiments: QC JC. Performed the experiments: QC. Analyzed the data: QC MLG. Wrote the paper: QC MLG JC.
References
- 1.
WHO (2013) World Malaria Report 2013. World Health Organization, Geneva.
- 2.
WHO (2012) Disease surveillance for malaria elimination: an operational manual. World Health Organization, Geneva, Switzerland.
- 3. Schneider P, Bousema JT, Gouagna LC, Otieno S, van de Vegte-Bolmer M, et al. (2007) Submicroscopic Plasmodium falciparum gametocyte densities frequently result in mosquito infection. Am J Trop Med Hyg 76: 470–474.
- 4. Coleman RE, Kumpitak C, Ponlawat A, Maneechai N, Phunkitchar V, et al. (2004) Infectivity of asymptomatic Plasmodium-infected human populations to Anopheles dirus mosquitoes in western Thailand. J Med Entomol 41: 201–208.
- 5. Okell LC, Ghani AC, Lyons E, Drakeley CJ (2009) Submicroscopic infection in Plasmodium falciparum-endemic populations: a systematic review and meta-analysis. J Infect Dis 200: 1509–1517.
- 6. Okell LC, Bousema T, Griffin JT, Ouedraogo AL, Ghani AC, et al. (2012) Factors determining the occurrence of submicroscopic malaria infections and their relevance for control. Nat Commun 3: 1237.
- 7. Roper C, Elhassan IM, Hviid L, Giha H, Richardson W, et al. (1996) Detection of very low level Plasmodium falciparum infections using the nested polymerase chain reaction and a reassessment of the epidemiology of unstable malaria in Sudan. Am J Trop Med Hyg 54: 325–331.
- 8. Sattabongkot J, Maneechai N, Phunkitchar V, Eikarat N, Khuntirat B, et al. (2003) Comparison of artificial membrane feeding with direct skin feeding to estimate the infectiousness of Plasmodium vivax gametocyte carriers to mosquitoes. Am J Trop Med Hyg 69: 529–535.
- 9. Collins WE, Jeffery GM, Roberts JM (2004) A retrospective examination of the effect of fever and microgametocyte count on mosquito infection on humans infected with Plasmodium vivax.. Am J Trop Med Hyg 70: 638–641.
- 10. Battle KE, Karhunen MS, Bhatt S, Gething PW, Howes RE, et al. (2014) Geographical variation in Plasmodium vivax relapse. Malar J 13: 144.
- 11. Fancony C, Gamboa D, Sebastiao Y, Hallett R, Sutherland C, et al. (2012) Various pfcrt and pfmdr1 genotypes of Plasmodium falciparum cocirculate with P. malariae, P. ovale spp., and P. vivax in northern Angola. Antimicrob Agents Chemother 56: 5271–5277.
- 12. Starzengruber P, Fuehrer HP, Ley B, Thriemer K, Swoboda P, et al. (2014) High prevalence of asymptomatic malaria in south-eastern Bangladesh. Malar J 13: 16.
- 13. Steenkeste N, Incardona S, Chy S, Duval L, Ekala MT, et al. (2009) Towards high-throughput molecular detection of Plasmodium: new approaches and molecular markers. Malar J 8: 86.
- 14. Steenkeste N, Rogers WO, Okell L, Jeanne I, Incardona S, et al. (2010) Sub-microscopic malaria cases and mixed malaria infection in a remote area of high malaria endemicity in Rattanakiri province, Cambodia: implication for malaria elimination. Malar J 9: 108.
- 15. Kaisar MM, Supali T, Wiria AE, Hamid F, Wammes LJ, et al. (2013) Epidemiology of Plasmodium infections in Flores Island, Indonesia using real-time PCR. Malar J 12: 169.
- 16. Asih PB, Rozi IE, Herdiana, Pratama NR, Hidayati AP, et al. (2012) The baseline distribution of malaria in the initial phase of elimination in Sabang Municipality, Aceh Province, Indonesia. Malar J 11: 291.
- 17. Khaminsou N, Kritpetcharat O, Daduang J, Kritpetcharat P (2008) A survey of malarial infection in endemic areas of Savannakhet province, Lao PDR and comparative diagnostic efficiencies of Giemsa staining, acridine orange staining, and semi-nested multiplex PCR. Parasitol Int 57: 143–149.
- 18. Singh B, Cox SJ, Miller AO, Abdullah MS, Snounou G, et al. (1996) Detection of malaria in Malaysia by nested polymerase chain reaction amplification of dried blood spots on filter papers. Trans R Soc Trop Med Hyg 90: 519–521.
- 19. Kritsiriwuthinan K, Ngrenngarmlert W (2011) Molecular screening of Plasmodium infections among migrant workers in Thailand. J Vector Borne Dis 48: 214–218.
- 20. Alves FP, Durlacher RR, Menezes MJ, Krieger H, Silva LH, et al. (2002) High prevalence of asymptomatic Plasmodium vivax and Plasmodium falciparum infections in native Amazonian populations. Am J Trop Med Hyg 66: 641–648.
- 21. Souza CR, Carvalho TA, Amaral RC, Cunha LS, Cunha MG, et al. (2012) Prevalence of Plasmodium falciparum and P. vivax in an area of transmission located in Para State, Brazil, determined by amplification of mtDNA using a real-time PCR assay. Genet Mol Res 11: 3409–3413.
- 22. Suarez-Mutis MC, Cuervo P, Leoratti FM, Moraes-Avila SL, Ferreira AW, et al. (2007) Cross sectional study reveals a high percentage of asymptomatic Plasmodium vivax infection in the Amazon Rio Negro area, Brazil. Rev Inst Med Trop Sao Paulo 49: 159–164.
- 23. Ladeia-Andrade S, Ferreira MU, de Carvalho ME, Curado I, Coura JR (2009) Age-dependent acquisition of protective immunity to malaria in riverine populations of the Amazon Basin of Brazil. Am J Trop Med Hyg 80: 452–459.
- 24. Katsuragawa TH, Gil LH, Tada MS, de Almeida e Silva A, Costa JD, et al. (2010) The dynamics of transmission and spatial distribution of malaria in riverside areas of Porto Velho, Rondonia, in the Amazon region of Brazil. PLoS One 5: e9245.
- 25. Cerutti C Jr, Boulos M, Coutinho AF, Hatab Mdo C, Falqueto A, et al. (2007) Epidemiologic aspects of the malaria transmission cycle in an area of very low incidence in Brazil. Malar J 6: 33.
- 26. Versiani FG, Almeida ME, Melo GC, Versiani FO, Orlandi PP, et al. (2013) High levels of IgG3 anti ICB2-5 in Plasmodium vivax-infected individuals who did not develop symptoms. Malar J 12: 294.
- 27. Roshanravan B, Kari E, Gilman RH, Cabrera L, Lee E, et al. (2003) Endemic malaria in the Peruvian Amazon region of Iquitos. Am J Trop Med Hyg 69: 45–52.
- 28. Rosas-Aguirre A, Llanos-Cuentas A, Speybroeck N, Cook J, Contreras-Mancilla J, et al. (2013) Assessing malaria transmission in a low endemicity area of north-western Peru. Malar J 12: 339.
- 29. Rodulfo H, De Donato M, Mora R, Gonzalez L, Contreras CE (2007) Comparison of the diagnosis of malaria by microscopy, immunochromatography and PCR in endemic areas of Venezuela. Braz J Med Biol Res 40: 535–543.
- 30. Mehlotra RK, Lorry K, Kastens W, Miller SM, Alpers MP, et al. (2000) Random distribution of mixed species malaria infections in Papua New Guinea. Am J Trop Med Hyg 62: 225–231.
- 31. Cole-Tobian JL, Cortes A, Baisor M, Kastens W, Xainli J, et al. (2002) Age-acquired immunity to a Plasmodium vivax invasion ligand, the duffy binding protein. J Infect Dis 186: 531–539.
- 32. Mueller I, Widmer S, Michel D, Maraga S, McNamara DT, et al. (2009) High sensitivity detection of Plasmodium species reveals positive correlations between infections of different species, shifts in age distribution and reduced local variation in Papua New Guinea. Malar J 8: 41.
- 33. Lin E, Kiniboro B, Gray L, Dobbie S, Robinson L, et al. (2010) Differential patterns of infection and disease with P. falciparum and P. vivax in young Papua New Guinean children. PLoS One 5: e9047.
- 34. Kasehagen LJ, Mueller I, McNamara DT, Bockarie MJ, Kiniboro B, et al. (2006) Changing patterns of Plasmodium blood-stage infections in the Wosera region of Papua New Guinea monitored by light microscopy and high throughput PCR diagnosis. Am J Trop Med Hyg 75: 588–596.
- 35. Harris I, Sharrock WW, Bain LM, Gray KA, Bobogare A, et al. (2010) A large proportion of asymptomatic Plasmodium infections with low and sub-microscopic parasite densities in the low transmission setting of Temotu Province, Solomon Islands: challenges for malaria diagnostics in an elimination setting. Malar J 9: 254.
- 36. Lekweiry KM, Abdallahi MO, Ba H, Arnathau C, Durand P, et al. (2009) Preliminary study of malaria incidence in Nouakchott, Mauritania. Malar J 8: 92.
- 37. Lekweiry KM, Basco LK, Salem MS, Hafid JE, Marin-Jauffre A, et al. (2011) Malaria prevalence and morbidity among children reporting at health facilities in Nouakchott, Mauritania. Trans R Soc Trop Med Hyg 105: 727–733.
- 38. Charlwood JD, Qassim M, Elnsur EI, Donnelly M, Petrarca V, et al. (2001) The impact of indoor residual spraying with malathion on malaria in refugee camps in eastern Sudan. Acta Trop 80: 1–8.
- 39. Singh N, Shukla MM, Shukla MK, Mehra RK, Sharma S, et al. (2010) Field and laboratory comparative evaluation of rapid malaria diagnostic tests versus traditional and molecular techniques in India. Malar J 9: 191.
- 40. Kim TS, Kim HH, Lee SS, Na BK, Lin K, et al. (2010) Prevalence of Plasmodium vivax VK210 and VK247 subtype in Myanmar. Malar J 9: 195.
- 41. Khattak AA, Venkatesan M, Nadeem MF, Satti HS, Yaqoob A, et al. (2013) Prevalence and distribution of human Plasmodium infection in Pakistan. Malar J 12: 297.
- 42. Tirasophon W, Rajkulchai P, Ponglikitmongkol M, Wilairat P, Boonsaeng V, et al. (1994) A highly sensitive, rapid, and simple polymerase chain reaction-based method to detect human malaria (Plasmodium falciparum and Plasmodium vivax) in blood samples. Am J Trop Med Hyg 51: 308–313.
- 43. Kuamsab N, Putaporntip C, Pattanawong U, Jongwutiwes S (2012) Simultaneous detection of Plasmodium vivax and Plasmodium falciparum gametocytes in clinical isolates by multiplex-nested RT-PCR. Malar J 11: 190.
- 44. Calderaro A, Gorrini C, Peruzzi S, Piccolo G, Dettori G, et al. (2008) An 8-year survey on the occurrence of imported malaria in a nonendemic area by microscopy and molecular assays. Diagn Microbiol Infect Dis 61: 434–439.
- 45. Peruzzi S, Gorrini C, Piccolo G, Calderaro A, Dettori G, et al. (2007) Prevalence of imported malaria in Parma during 2005–2006. Acta Biomed 78: 170–175.
- 46. Rubio JM, Benito A, Berzosa PJ, Roche J, Puente S, et al. (1999) Usefulness of seminested multiplex PCR in surveillance of imported malaria in Spain. J Clin Microbiol 37: 3260–3264.
- 47. Andrade BB, Reis-Filho A, Barros AM, Souza-Neto SM, Nogueira LL, et al. (2010) Towards a precise test for malaria diagnosis in the Brazilian Amazon: comparison among field microscopy, a rapid diagnostic test, nested PCR, and a computational expert system based on artificial neural networks. Malar J 9: 117.
- 48. Durrheim DN, Becker PJ, Billinghurst K (1997) Diagnostic disagreement—the lessons learnt from malaria diagnosis in Mpumalanga [letter]. S AfrMedJ 87: 1016.
- 49. Kachur SP, Nicolas E, Jean FV, Benitez A, Bloland PB, et al. (1998) Prevalence of malaria parasitemia and accuracy of microscopic diagnosis in Haiti, October 1995. Rev Panam Salud Publica 3: 35–39.
- 50. McKenzie FE, Sirichaisinthop J, Miller RS, Gasser RA Jr, Wongsrichanalai C (2003) Dependence of malaria detection and species diagnosis by microscopy on parasite density. Am J Trop Med Hyg 69: 372–376.
- 51. Ohrt C, Purnomo , Sutamihardja MA, Tang D, Kain KC (2002) Impact of microscopy error on estimates of protective efficacy in malaria-prevention trials. J Infect Dis 186: 540–546.
- 52. Milne LM, Kyi MS, Chiodini PL, Warhurst DC (1994) Accuracy of routine laboratory diagnosis of malaria in the United Kingdom. JClinPathol 47: 740–742.
- 53. Bell D, Go R, Miguel C, Walker J, Cacal L, et al. (2001) Diagnosis of malaria in a remote area of the Philippines: comparison of techniques and their acceptance by health workers and the community. Bull World Health Organ 79: 933–941.
- 54. Coleman RE, Maneechai N, Rachaphaew N, Kumpitak C, Miller RS, et al. (2002) Comparison of field and expert laboratory microscopy for active surveillance for asymptomatic Plasmodium falciparum and Plasmodium vivax in western Thailand. Am J Trop Med Hyg 67: 141–144.
- 55. Anderson TJ, Haubold B, Williams JT, Estrada-Franco JG, Richardson L, et al. (2000) Microsatellite markers reveal a spectrum of population structures in the malaria parasite Plasmodium falciparum. Mol Biol Evol 17: 1467–1482.
- 56. Branch OH, Sutton PL, Barnes C, Castro JC, Hussin J, et al. (2011) Plasmodium falciparum genetic diversity maintained and amplified over 5 years of a low transmission endemic in the Peruvian Amazon. Mol Biol Evol 28: 1973–1986.
- 57. Ord RL, Tami A, Sutherland CJ (2008) ama1 genes of sympatric Plasmodium vivax and P. falciparum from Venezuela differ significantly in genetic diversity and recombination frequency. PLoS One 3: e3366.
- 58. Gray KA, Dowd S, Bain L, Bobogare A, Wini L, et al. (2013) Population genetics of Plasmodium falciparum and Plasmodium vivax and asymptomatic malaria in Temotu Province, Solomon Islands. Malar J 12: 429.
- 59. Gravenor MB, Kwiatkowski D (1998) An analysis of the temperature effects of fever on the intra-host population dynamics of Plasmodium falciparum. Parasitology 117: 97–105.
- 60. Kwiatkowski D, Nowak M (1991) Periodic and chaotic host-parasite interactions in human malaria. Proc Natl Acad Sci U S A 88: 5111–5113.
- 61.
WHO (2010) Guidelines for the treatment of malaria — 2nd edition. World Health Organization, Geneva.
- 62. Poirot E, Skarbinski J, Sinclair D, Kachur SP, Slutsker L et al. (2013) Mass drug administration for malaria. Cochrane Database Syst Rev.