Asymptomatic Vivax and Falciparum Parasitaemia with Helminth Co-Infection: Major Risk Factors for Anaemia in Early Life

Background Anaemia in children under five years old is associated with poor health, growth and developmental outcomes. In Papua, Indonesia, where the burden of anaemia in infants is high, we conducted a community survey to assess the association between Plasmodium infection, helminth carriage and the risk of anaemia. Methods A cross sectional household survey was carried out between April and July 2013 in 16 villages in the District of Mimika using a multistage sampling procedure. A total of 629 children aged 1–59 months from 800 households were included in the study. Demographic, symptom and anthropometry data were recorded using a standardized questionnaire. Blood and stool samples were collected for examination. Results Of the 533 children with blood film examination, 8.8% (47) had P. vivax parasitaemia and 3.9% (21) had P. falciparum; the majority of children with malaria were asymptomatic (94.4%, 68/72). Soil transmitted helminth (STH) infection was present in 43% (105/269) of children assessed; those with STH were at significantly greater risk of P. vivax parasitaemia compared to those without STH (OR = 3.7 [95%CI 1.5–9.2], p = 0.004). Anaemia (Hb<10 g/dl) was present in 24.5% (122/497) of children and associated with P. vivax parasitaemia (OR = 2.9 [95%CI, 1.7–4.9], p = 0.001), P. falciparum parasitaemia (OR = 4.3 [95%CI, 2.0–9.4], p<0.001), hookworm carriage (OR = 2.6 [95%CI, 1.2–5.8], p = 0.026), Plasmodium–helminth coinfection (OR 4.0 [95%CI, 1.4–11.3], p = 0.008) and severe stunting (OR = 1.9 ([95%CI, 1.1–3.3], p = 0.012). Conclusions Asymptomatic P. vivax and P. falciparum infections and hookworm all contribute to risk of paediatric anaemia in coendemic areas and should be targeted with prevention and treatment programs. The relationship between helminth infections and the increased risk of P. vivax parasitaemia should be explored prospectively.


Introduction
Anaemia remains a critical determinant of global morbidity and mortality in children under five years old [1][2][3]. Whilst mild anaemia is usually asymptomatic and may go unrecognized by both patients and healthcare providers [4], it has been associated with impaired immune function [5][6][7][8]. Anaemic individuals in poorly resourced communities are vulnerable to a cycle driven by low host immunity, an increased susceptibility to malnutrition and infection, which give rise to further haemopoietic suppression and haemolysis. As the hemoglobin concentration falls below 8g/dl the risk of mortality rises [1,9]. In those who do not succumb, the consequences of severe anaemia can be long-lasting, including cognitive impairment, stunting and delays in motor development [10][11][12].
The aetiology of anaemia in malaria-endemic areas is multifactorial. Risk factors include parasite-induced haemolysis and haemopoietic suppression, iron deficiency, poor nutritional intake, recurrent bacterial infections and soil transmitted helminthiases [13] [2,9]. Whilst the health consequences of asymptomatic falciparum malaria and its interplay with other causes of anaemia has been relatively well defined in children in P. falciparum endemic areas of Africa [4,[14][15][16], the interplay between P. vivax infection and stool transmitted helminth (STH) infections is poorly understood with little known about the factors associated with anaemia attributable to P. vivax [17].
Observational studies in Papua province in eastern Indonesia, have demonstrated that severe anaemia (Hb<5 g/dl) is present in almost 20% of young children attending an inpatient healthcare facility; approximately a third of this anaemia attributable to malaria [3,9]. Papua, is a malaria endemic region unusual in having high prevalence of antimalarial drug resistance in both P. falciparum and P. vivax [18,19]. Recurrent clinical episodes and sustained parasitaemia are key contributing factors and occur because of treatment failure (recrudescence), reinfection and relapsing infections from the dormant liver stages of P. vivax.
The prevalence of STH in Papua has not been well characterized, nor its role in P. vivax coinfections and in the aetiology of anaemia. In the current cross sectional survey, the relationship between Plasmodium infection, nutritional status, helminth coinfection and anaemia was assessed to provide the evidence base for suitable intervention strategies. almost 90% of the population located in the lowlands, which is largely forested with a climate that varies little throughout the year [20]. There are 12 sub-districts and 79 villages in an area of 19,602 km 2 , with the largest three sub-districts reachable by land from the capital city (Timika).
Malaria in the region is restricted to lowland areas where it is associated with three mosquito vectors: Anopheles koliensis, An. farauti and An. punctulatus. The area has unstable malaria transmission with an estimated annual incidence of parasitaemia of 450 per 1000 population in 2013, divided 60:40 between P. falciparum and P. vivax infections [21]. Clinical trials in 2004 revealed extremely high antimalarial treatment failure, with 65% of patient having recurrent parasitaemia 28 days after chloroquine (CQ) monotherapy for P. vivax and 48% on day 42 after CQ plus sulfadoxine-pyrimethamine (SP) for P. falciparum [18]. As previously mentioned, the first line treatment policy for uncomplicated malaria in Mimika was changed in March 2006 to dihydroartemisinin-piperaquine (DHP) for malaria due to any Plasmodium species [22]. Insecticide treated nets are used by approximately 52% of the population [21].
Soil transmitted helminthiases (STHs) are prevalent in malaria-endemic areas, but precise data in this region are lacking. There is no specific program for deworming or iron supplementation in young children in the district.

Study population
Children aged less than 5 years old were enrolled in this study as part of a larger household survey that included all consenting members of the households. This age group accounts for 15% of the local population [20]. The under five year mortality rate in this region is 115/1000 live births with a neonatal mortality rate of 27/1000 live births [23]. Malaria is the main cause of outpatient and inpatient visits to the local hospital in under fives, followed by diarrhea and lower respiratory tract infections [24].
The ethnicity of the population can be broadly categorized into highland and lowland Papuans and non-Papuans.

Study design
A cross sectional survey was carried out between April and July 2013 in 16 villages in 3 sub-districts in the District of Mimika to evaluate the impact of malaria treatment policy change to malaria prevalence as described previously [19]. Surveys were conducted before and after treatment policy change in March 2006. The first survey was conducted in 2005 and the second one in 2013.
The survey sample size was powered to detect a 40% reduction in malaria prevalence assuming an estimated incidence of malaria of 500 episodes/1000 people/year; a total of 4000 participants were required. In the District of Mimika, each household consists of approximately 5 persons and this amounted to 800 households. Children under-fives in the households were included in the study.
A multistage sampling procedure was applied to identify clusters of 25 households. Firstly the three largest sub-districts were chosen purposively from the 12 sub-districts based on their accessibility and inclusion of the residence of the majority (80%) of the population. The proportions of ethnicity and socioeconomic strata was similar between this three subdistricts [20]. Secondly, the number of clusters in each sub-district was apportioned according to the relative populations of the subdistricts. The houses were then assigned numbers according to their geographical location and clusters of 25 houses were identified randomly according to WHO guidelines with the first house identified by random allocation [25].
A total of 32 clusters of 25 randomly selected households were included in the survey providing a total of 800 households. Household members were defined as people who lived under one roof, ate from one kitchen and who had resided in the study area for at least 6 months (except for infants younger than 6 months).
For each household member, sociodemographic information and history of fever were recorded using a standardised questionnaire. Socioeconomic status was defined according to the national standard of minimal employer's wage in the region. Those present at the time of the survey had their weight, height and temperature recorded and a finger prick sample of blood taken for blood film examination and haemoglobin measurement. Patients with microscopically confirmed malaria were treated with DHP [26]. Stool samples were collected from all children aged less than 5 years old. The parent/guardian was given a sealed plastic container with stick to collect small amount of stool sample on the first visit for collection on the next day in the morning.
Haemoglobin concentration was determined by electronic coulter counter (Coulter JT™, USA). Malaria was defined as the presence of peripheral asexual parasitaemia by microscopy irrespective of clinical signs or Plasmodium species. Two trained research microscopists examined the slides independently. Two different microscopists examined the stool samples using the Katz Kato method and recorded both the species present and concentration of eggs per gram of stool according to the guidelines [27,28]. The final result of parasite and egg counts was the average of the two different counts. If there was a discrepancy in the species diagnosis or a difference in the parasite and egg count greater than 50%, an independent third reader reexamined the slide and final result made according to consensus.

Definitions
An 'infant' was defined as a child aged less than 1 year old and a "young child" if aged 1 to <5 years. Children under-five included both infants and young children. Anaemia was defined as a hemoglobin concentration less than 10 g/dl. Nutritional status was assessed according to the WHO 2007 Standard Reference [29]. Underweight and severe underweight were defined as weight for age Z-scores below-2SD and-3 SD respectively. Similar categories were applied to wasting-severe wasting and stunting-severe stunting by examining Z-scores of height/length for weight and height/length for age respectively.

Statistical analysis
Questionnaire and laboratory data were entered into EpiData 3.02 software (EpiData Association, Odense, Denmark). Data were analysed using SPSS vs 20.0 for windows software (IBM SPSS Statistics). Normally distributed data were compared by Student's t-test. Data not conforming to a normal distribution were compared by the Mann-Whitney U test. Categorical data were compared by using odds ratios (OR with 95% confidence intervals), the chi squared test with Yates' correction or by Fisher's exact test. Multiple logistic regression was used to analyse independent risk factors for anaemia by entering all significant risk factors with p value <0.05 in univariate analysis.

Ethical Approval
Ethical approval for this study was obtained from the Medical and Health Research Ethics Committee, Faculty of Medicine, Gadjah Mada University, Yogyakarta, Indonesia (KE/FK/39/ EC), Menzies School of Health Research, Darwin, Australia (HREC 10-1434) and Oxford Tropical Research Ethics Committee (44-1). Written consent was obtained from the head of the household, and included consent for all members of the family to participate in the study, including children aged less than 16 years old.

Results
Between April and July 2013 a total of 629 children aged 1 to 59 months old (median age 24 months, range 1-59 months) were enrolled from 800 households; see Fig 1. Infants accounted for 17% (108) of children under-five. Characteristics of the study participants are presented in Table 1. Of the 531 children with available weight data and the 507 children with length and height data, 76 (14.3%) were classified as underweight, 47 (9.3%) as wasted, and 16.7% (85/ 507) stunted; these risks were similar between infants and young children and were not associated with ethnicity or socioeconomic status (Table 1).

Peripheral Parasitaemia by microscopy
Blood film examination was undertaken in 85% (533) of children of whom 8.8% (47)
The prevalence of helminthiasis was 19.4% (7/36) in infants compared to 40.8% (95/233) in children aged 1 to 5 years (Tables 1 and 3). The egg burden did not significantly differ with age, however the number of infections in the infant group was small (   Table 5. Of all the nutritional status indices, only severe stunting was associated with an increased risk of anaemia (OR = 1.9 [95%CI, 1.1-3.3] compared to well-nourished children, p = 0.012). The prevalence of anaemia was 34.1% (28/82) in children with severe stunting and 21.1% (65/ 308) in those with normal height according to age; p = 0.02.

Discussion
In Papua, Indonesia, anaemia exerts a significant morbidity particularly in young children. Previous hospital based studies have highlighted the importance of undetected, untreated or inadequately treated malaria as a key contributor to reduction in haemoglobin concentrations, [3,9]. The cross sectional study presented here provides additional insights into the importance of Plasmodium infection and the importance of soil transmitted helminths. Plasmodium infection was present in 12% of children under 5 years old, most of whom were asymptomatic, with P. vivax the dominant species of infection. Helminth carriage was common and when present with malaria doubled the risk of anaemia. The aetiology of anaemia from malaria and helminth carriage is well described and is attributable to a combination of chronic blood loss, haemolysis and haemopoietic suppression. Associated nutritional deficiency is also likely to play a key role. Although in our study general measures of stunting and wasting were not significant risk factors, in the subset of individuals tested, microcytosis was present in almost 75%, suggesting a potential contribution of iron deficiency. Recent studies have demonstrated that increased hepcidin level and associated impaired intestinal iron absorption may play an important role in the development of anaemia in children with asymptomatic Plasmodium infection and may also play a role in P. vivax infections [30]. The high proportion of asymptomatic malaria in young children in this region suggests that clinical immunity develops at an early age in both P. falciparum and P. vivax infections [3,31],   although parasitaemias were low and it is plausible that the survey detected an early phase of infection prior to development of symptoms. The burden of P. vivax is particularly apparent in children under 5 years old, with the prevalence (8.8%) more than twice that of P. falciparum (3.9%). Plasmodium vivax peripheral parasitaemias were lower than in P. falciparum infections and children with P. vivax more likely to be asymptomatic. However the relapsing nature of P. vivax infections is an important determinant of anaemia in the first year of life [3] and is a major challenge for healthcare providers and clinicians since current National guidelines preclude the use of primaquine for infants aged less than 1 year [32]. In addition, the 2015 WHO guidelines only suggests primaquine use in in infants older than 6 months old [33]. Soil transmitted helminths (STHs) were present in almost half of the children by 5 years of age, and of these, almost half had multiple infestations. Stool sampling was undertaken in less than half of the children. The characteristics of those with and without stool samples (age, sex and parental socioeconomic status) was similar except for ethnicity, with 60% of Papuans were without stool sampling compared to 51% in non Papuan children. Papuan ethnicity was a major independent risk factor for helminthiasis raising the prevalence 2 to 4 fold, hence our analysis is likely to have underestimated the true burden of helminth infection.
The presence of hookworm was particularly important, increasing the risk of anaemia 2.6 fold. Once again the true burden is likely to be actually higher since Kato Katz has low sensitivity in diagnosing hookworm eggs [34]. Hookworm infestation is more prevalent in later childhood and early adulthood and is associated with chronic intestinal blood loss [14] [35]. Although less common in infants than older children, the prevalence of hookworm rose rapidly in the first year of life. Although our study did not quantify hookworm carriage in older children, the presence of this parasite at such an early age is interesting and may reflect earlier exposure to soil [35].
In Papua, children with malaria and STH co-infection were four times as likely to have anaemia compared to those having mono-infection with either pathogen, and at almost nine fold greater risk of severe anaemia. Although the frequency of co-infection was relatively low (7.1%), the presence of co-infection and its association with anaemia is likely to have a significant detrimental effect on the child's health and development [14]. Similar findings were observed in Africa, where P. falciparum and hookworm comorbidity in preschool children was Anaemia, Malaria and Helminth Infection in Children under Five associated with lower haemoglobin concentrations compared to those having single infection [36]. On the other hand, school aged children with P. falciparum infections and hookworm infestation had lower risk of developing iron deficiency anaemia [37]. Although not generally recognized as a major cause of anaemia, in our study ascaris and trichuris as well as hookworm were associated with twice the risk of anaemia, although after controlling for ethnicity only hookworm remained a significant factor. Dysregulation of cellular immune responses in STH infection [38][39][40] may down regulate protective immune responses to Plasmodium spp. in a complex way, increasing susceptibility to falciparum malaria without resulting in severe symptoms [36,41,42]. Although there are few studies from vivax-endemic areas, the interplay with P. vivax infection may be different. Heavy ascariasis infestation reduced the risk of symptomatic P. vivax malaria in one study of children aged 2-14 years old [43]. In another, STH infection protected against a further reduction in haemoglobin during acute vivax malaria [44]. Studies of immune responses, STH and malaria remain inconclusive and are not informative for P. vivax since most have been conducted in sub-Saharan Africa where P. vivax is not endemic [42,[45][46][47][48][49].
The sample size of our study was designed to evaluate the impact of antimalarial treatment policy by comparison with a prior survey conducted in 20015. The extensive data collected in the second survey enabled us to define the effect of asymptomatic malaria and STH infections to anaemia in children under-fives. The sample size of this study was sufficient to test the association of malaria, STH and anaemia [24]. However our study has a number of limitations. Firstly, the relative contribution of recurrent or chronic Plasmodium infection and STHs to the development of anaemia could not be examined in this cross sectional study. However in this population, individuals with a history of fever in the previous 30 days is a useful indicator of symptomatic malaria and in the cross sectional surveys was associated with a 3.6 fold higher risk of anaemia, suggesting a likely contribution of recurrent malaria. Secondly, iron status was not quantified so the contribution of iron deficiency could not be addressed. However in the subset of patients in whom data were available, three quarters of children with anaemia had microcytic erythrocytes suggesting iron deficiency and/or infections in the underlying aetiology.
In conclusion, asymptomatic Plasmodium infection and malaria/helminth coinfection are important risk factors for anaemia in young children. STHs appear to increase a child's risk of P. vivax parasitaemia. Operational studies are needed to define the effectiveness of combined malaria and STH control programs to prevent their combined contribution to anaemia in early life and associated morbidity, mortality and cognitive impairment. Our study suggests that deworming has potential to reduce P. vivax infections. In the case of malaria, the risks and benefits of malaria screening or preventive treatment in infancy in areas endemic for P. falciparum and P. vivax warrants careful examination [50].

(XLSX)
Research and Collaboration), Dr. Yodi Mahendradhata from the Faculty of Medicine Universitas Gadjah Mada, Yogyakarta and Dr Ella Curry for their excellent support to this study. We thank the field and laboratory staff of the Papuan Health and Community Development Foundation and Steven Kho for logistical and technical support.