Background
Malaria is still a major cause of morbidity and mortality in India, although the numbers are decreasing. In 2018, the National Vector Borne Disease Control Programme (NVBDCP) estimated that approximately 500,000 people suffered from malaria (63%
Plasmodium falciparum), and less than 100 persons died [
1]. These numbers are likely an underestimate; the World Malaria Report estimated that only 16% of cases may be detected by the Indian malaria surveillance system [
2]. Malaria in India is complex, with multiple
Plasmodium species and
Anopheles vectors, and a wide variety of endemic settings [
3].
Plasmodium vivax can be detected throughout the country, whereas
P. falciparum is more common in the Eastern and North-Eastern states. The major malaria vector is
Anopheles culicifacies, estimated to be associated with 60–65% of the malaria disease burden in India [
3].
Anopheles stephensi is present in urban and peri-urban areas, and has been associated with serious malaria outbreaks in South India [
4], whereas
Anopheles fluviatilis is present in the forests and foothills, and
Anopheles minimus mainly in the North-Eastern states [
3]. The diverse malaria settings make the goal of malaria elimination challenging; one size does not fit all, and multiple localized strategies are needed, a stance supported by the National Framework for Malaria Elimination in India [
5].
In the past in the absence of a diagnostic test, it was common to diagnose malaria based only on symptoms. Several studies have evaluated if specific symptoms or combinations of symptoms were reliable to diagnose malaria [
6‐
8]; however, the symptoms of malaria are not very specific and can be caused by many diseases. In addition, malaria can present in unusual ways or can be part of co-morbidities [
9‐
11]. Microscopy used to be the other main means of diagnosing malaria. More recent methods include malaria rapid diagnostic tests (RDTs) and molecular methods, such as polymerase chain reaction (PCR). RDTs are now widely used by local NVBDCP clinics and by community health workers (ASHAs) in India. Malaria testing using PCR requires expensive equipment and reagents as well as trained staff, which limits its use in local clinics. The greater sensitivity of PCR for detecting
Plasmodium parasites has resulted in the introduction of the terms ‘submicroscopic’ and ‘subpatent’ malaria, referring to cases where the number of malaria parasites is below the detection level of microscopy or RDT [
12,
13]. Surveys have repeatedly shown that not all cases of malaria present with symptoms; this phenomenon was well-known in African countries in areas of high malaria transmission intensity but is now also documented in areas with low malaria transmission and for both species of malaria parasites common in India [
14]. Asymptomatic infections are important as a hidden reservoir of
Plasmodium parasites [
14], and can contribute to continued transmission. In addition, asymptomatic infections can still have subtle adverse effects on health [
15,
16].
Symptoms of malaria among the general population can differ by region, depending on prevailing diseases [
17]. A published study from the National Institutes of Health-funded International Center of Excellence for Malaria Research describing cross-sectional malaria surveys conducted at three sites in India included a detailed questionnaire on symptoms [
3,
18]. Here this questionnaire data was used to assess the distribution of symptoms by region, age, and gender, and the association of symptoms with malaria. The goal was to determine the proportion of malaria associated with fever (documented fever or a history of fever in the past 48 h), or other symptoms (without fever), and the proportion of truly asymptomatic infections in these different regions.
Discussion
India aims to eliminate malaria [
5]. To do so it is important for all symptomatic and, if feasible, asymptomatic cases to be identified and treated. However, malaria in India is heterogeneous, and in addition, many diseases present with symptoms similar to it [
11]. In the three regions of India with different malaria endemicities studied here, malaria prevalence was low (detected by PCR: 2.6%, 7.0% and 7.7% in Chennai, Nadiad and Rourkela, respectively). Fever was the most common symptom of malaria; chills, aches, and headache were common symptoms reported among persons without fever but with a positive malaria test. There were considerable regional differences in (truly) asymptomatic malaria: 7.1% of all positive malaria cases detected by PCR in Nadiad, 39.6% in Rourkela and 66.7% in Chennai. In Rourkela, asymptomatic infection was associated with young age and male gender as detected by microscopy, and with season as detected by PCR.
Different patterns of general symptoms were observed, with the lowest number of any symptoms in Chennai, and the highest number of any symptoms in Nadiad. Patterns of symptoms were higher in the younger age groups and were more common in the dry season in Nadiad, whereas in Rourkela they were more common in the rainy season. Only in Rourkela were gender differences observed for some symptoms (headaches and aches more common among women, chills more common among men). These patterns in symptoms by region indicates that the burden of disease in each region is different. It has been reported that cultural factors can contribute to differences in symptom experiences and health-seeking behavior [
26]. Since several other diseases including influenza, dengue, and chikungunya have overlapping symptoms and all are prevalent in our study areas, it may be hard to differentiate between them [
27].
It is not always apparent whether asymptomatic infections will develop into symptomatic malaria cases and be cleared by treatment with anti-malarial drugs or will be cleared by the person’s immune system without intervention. Submicroscopic asymptomatic parasite levels can persist after treatment, if parasites are drug resistant or if inadequate doses of anti-malarials have been used. An association with recent treatment was not detected in this analysis, but in Chennai, asymptomatic infections (as detected by RDT) were associated with an infection in the past year. A recent study in a low endemicity area in Vietnam followed 356 persons with asymptomatic subpatent infections; 61% remained afebrile during the 24-month follow up period. They noted a median duration of asymptomatic submicroscopic
P. falciparum infection of 2 months, with variable parasite densities; for
P. vivax this was 6 months [
28]. Like the study presented here, parasite densities among febrile participants with
P. falciparum were higher than among non-febrile participants [
28]. In a study in Thailand, 80.5% of the 41 persons with malaria detected by PCR (prevalence 3.0%) had no complaints 5 months before, during or after the survey (follow up 6–15 months with weekly fever surveillance), three (7.3%) had fever at presentation, and five (12.2%) developed fever 1–4 months after presentation [
29]. A relationship between asymptomatic infections with anaemia in any of the sites was not apparent, but in Rourkela a significantly lower haemoglobin level (but not anaemia) was detected among persons with asymptomatic parasites.
Although malaria prevalence was similar in Nadiad and Rourkela, the proportion of asymptomatic infections was very low in Nadiad compared to Rourkela, and almost all malaria in Nadiad was accompanied by symptoms. It has been suggested that the asymptomatic reservoir could be minimal when transmission has decreased over many years and people have lost immunity [
30] and this might be the case in Nadiad, where 5% of persons had a history of malaria in the past year. In Rourkela, 25% of the population had a history of malaria in the past year and
P. falciparum was the main species; it was the only site with considerable bed net use. Malaria is decreasing in Rourkela [
31], but the decrease in transmission may be more rapid than the loss of immunity, resulting in this significant reservoir of asymptomatic carriers; this may also be the case for Chennai [
30,
32]. Immunity is affected by previous exposure to malaria, age, virulence, and number of infecting strains [
16,
30,
33]. Only at one site, Rourkela, could factors affecting asymptomatic malaria be explored. As detected by microscopy and RDT, it was associated with young age, which was contrary to the expectations that older people have more immunity and will be able to better control malaria [
33,
34]. Alternatively, asymptomatic infections may be present in older age groups but less likely to be detected because of lower parasite densities as a result of the higher immunity [
34]. It has been suggested that asymptomatic infections may maintain the parasite during low transmission seasons [
30,
34]; in Rourkela, they were more common in the rainy season (as detected by PCR).
Plasmodium vivax infections were more likely to be asymptomatic compared to
P. falciparum infections; results in the literature concerning this have been conflicting [
33]. It is possible that the limited number of circulating parasite strains in low transmission settings induces tolerance among persons at risk of infection and re-exposure, resulting in a higher proportion of asymptomatic infections as has been suggested for Pacific Islands; a limited number of strains may be circulating in Chennai, but data are not yet available to examine this [
30,
35]. Use of the term “chronic infections” instead of “asymptomatic infections” has been proposed because of the potential adverse effects on health of asymptomatic infections [
15,
16]. Asymptomatic infections can be transmitted to
Anopheles mosquitoes [
33] although gametocytaemia was very low in our studies; e.g., in Rourkela, prevalence was similar in symptomatic and asymptomatic infections. It is not clear if total removal of asymptomatic malaria is needed to achieve elimination [
30,
36].
Asymptomatic malaria is often defined as a positive malaria test in the absence of documented fever or a history of fever [
37]. This study shows the importance of using an appropriate case-definition for asymptomatic infections: fever, chills, aches, and headaches identified all symptomatic malaria cases in Chennai, and most in Nadiad. However, symptoms cannot replace testing for the presence of
Plasmodium parasites, and the definition of symptomatic/asymptomatic infections is mainly useful in the surveillance context for the evaluation of malaria burden. In Rourkela, there remained a group of persons who had symptoms such as cough or backache and a positive
Plasmodium test. It is possible that these symptoms were unrelated to malaria but due to other diseases, given the higher prevalence of asymptomatic
Plasmodium infections in this area.
Conclusion
This current study reports on regional differences in general complaints at the three sites in India and on difference of malaria and asymptomatic infections by region. Asymptomatic infections are important because it can continue transmission, hinder elimination, and chronically affect people’s health. Strategies to reduce asymptomatic infections include “mass screen and treat” campaigns, mass drug administration, and reactive case management, and all these options can be combined with indoor residual spraying or ITN. Reactive case detection (RCD), the screening of persons in the households and neighborhoods of positive index cases, has been studied in Chennai and Nadiad but results were not encouraging [
38]. RCD is labor intensive, is likely to miss the subpatent and submicroscopic infections depending on the malaria test and may not be effective for
P. vivax where dormant hypnozoites in the liver can continue transmission. Mass drug treatment has been an alternative but may result in ineffective exposure of persons to adverse events of medication when the malaria prevalence is very low, and may select resistant parasites, especially when conducted in the dry season [
34]. Mass screening and treatment is an alternative option; however, disadvantages are similar to RCD. A focused approach may be more effective when clusters have been identified in place and time. In Nadiad, given that the majority of malarious persons present with symptoms and there is a very low prevalence of subpatent and submicroscopic infections, prompt case-detection and effective treatment may reduce malaria cases further. In Chennai, the prevalence in the community is low but the proportion of asymptomatic persons is relatively high and additional action may be needed. In Rourkela, focusing on “hotspots” and at-risk population groups may assist in further reduction of malaria, for example the malaria camp approach currently being used by Odisha State Government to reduce malaria as part of a Durgama Anchalare Malaria Nirakaran (DAMaN)
‘malaria control in inaccessible areas’ programme.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.