Re-Visiting Trichuris trichiura Intensity Thresholds Based on Anemia during Pregnancy

Background The intensity categories, or thresholds, currently used for Trichuris trichiura (ie. epg intensities of 1–999 (light); 1,000–9,999 epg (moderate), and ≥10,000 epg (heavy)) were developed in the 1980s, when there were little epidemiological data available on dose-response relationships. This study was undertaken to determine a threshold for T. trichiura-associated anemia in pregnant women and to describe the implications of this threshold in terms of the need for primary prevention and chemotherapeutic interventions. Methodology/Principal Findings In Iquitos, Peru, 935 pregnant women were tested for T. trichiura infection in their second trimester of pregnancy; were given daily iron supplements throughout their pregnancy; and had their blood hemoglobin levels measured in their third trimester of pregnancy. Women in the highest two T. trichiura intensity quintiles (601–1632 epg and ≥1633 epg) had significantly lower mean hemoglobin concentrations than the lowest quintile (0–24 epg). They also had a statistically significantly higher risk of anemia, with adjusted odds ratios of 1.67 (95% CI: 1.02, 2.62) and 1.73 (95% CI: 1.09, 2.74), respectively. Conclusions/Significance This analysis provides support for categorizing a T. trichiura infection ≥1,000 epg as ‘moderate’, as currently defined by the World Health Organization. Because this ‘moderate’ level of T. trichiura infection was found to be a significant risk factor for anemia in pregnant women, the intensity of Trichuris infection deemed to cause or aggravate anemia should no longer be restricted to the ‘heavy’ intensity category. It should now include both ‘heavy’ and ‘moderate’ intensities of Trichuris infection. Evidence-based deworming strategies targeting pregnant women or populations where anemia is of concern should be updated accordingly.


Introduction
The most recent comprehensive estimation of the prevalences of the soil-transmitted helminthiases (STH) documents a global prevalence of 17% for Trichuris trichiura infection, with approximately 800 million persons infected at any one time [1,2]. Community-wide prevalences are frequently over 30-40% and it is not uncommon to observe prevalences exceeding 80% in community sub-groups like school-age children and preschool-age children [3][4][5][6][7]. T. trichiura infections contribute to the STHattributable burden of disease by adversely affecting the growth and cognitive development of children and the health and productivity of adults [8,9]. Because of its co-occurrence with other infections, malnutrition and poverty, it also diminishes the economic potential, not only of the infected individual, but also of the family and community as well [10].
In 1987, an expert committee convened by the World Health organization (WHO) established infection intensity categories for STH, including T. trichiura, in order to inform the management of large-scale deworming programs [11]. T. trichiura infection was defined as light (1-999 epg) or heavy (.10,000 epg) [11]. These categories were based primarily on expert opinion and little doseresponse data from the field, and were described as ''arbitrary'' by this committee. [11]. A further category of 'moderate' (i.e. for epg counts between 1,000 and 9,999 epg) was subsequently added by WHO [12]. The original 1987 report had also mentioned that anemia attributable to T. trichiura infection reflected a 'very heavy worm burden' [11].
Since then, the association between T. trichiura (prevalence and intensity) and hemoglobin (Hb) levels or anemia, has been assessed in several epidemiologic studies mostly conducted in Africa and in Asia and of which the majority found no significant association [13][14][15][16][17][18]. However, four studies conducted in the Americas (Jamaica, Panama, Mexico and Peru) reported statistically significant associations [19][20][21][22]. In addition, T. trichiura infection has been associated with a lower increase of Hb in ironsupplemented pregnant women [22]. Mechanisms by which T. trichiura infection may cause anemia include ingestion of blood by the parasite, blood loss from parasite-induced lesions in the intestinal mucosa, and inflammatory responses such as tumor necrosis factor a (TNFa) leading to decreased appetite; the relative contributions of these factors being unknown [9].
Anemia is a major public health problem because it impairs the growth and cognitive development in children and because severe anemia increases the risk of maternal mortality. Its worldwide prevalence is estimated at 48.8% [23]. The importance of the cluster of STH to the global risk of anemia is relatively well known, but among helminth species, T. trichiura has received much less attention than hookworms.
The objectives of this study were to determine a threshold for T. trichiura-associated anemia in pregnant women, and to describe the implications of this threshold in terms of the need for primary prevention and chemotherapeutic interventions.

Ethics Statement
Ethics approval was obtained for the original RCT from the following review committees: Research Institute of the McGill University Health Centre (Canada), The ''Comite Institucional de Etica de la Universidad Peruana Cayetano Heredia'' (Peru); and the ''Comite Etica de la Direccion General de Salud de las personas del Ministerio de Salud de Peru'' (Peru). The research procedures followed were in accordance with the ethical standards of these three ethics committees and with the Helsinki Declaration. Written informed consent was obtained from all women.
The data source for this study originated from a randomized controlled trial on mebendazole during pregnancy and its effect on birth weight which had been conducted in the highly STHendemic Amazon area of Peru whose methods have been described elsewhere [24]. Briefly, 1,042 pregnant women were recruited in their second trimester and randomly assigned to receive either a single dose of 500 mg mebendazole or a placebo. Women in both groups received daily iron supplements throughout their pregnancy. At enrolment (second trimester) and again in the third trimester, blood and stool specimens were collected from participants for hemoglobin (Hb) ascertainment by HemoCue and for STH determination by the Kato-Katz method. There was no statistically significant difference between intervention groups in the prevalence of anemia or in mean hemoglobin levels in the third trimester. However, women having Trichuris trichiura infection in the second trimester were at a higher risk of anemia in their third trimester [22].
To determine a threshold for the effect of T. trichiura infection intensity on hemoglobin and anemia, the 935 mothers for whom complete information was available (i.e. on helminth infection and hemoglobin level in both the 2nd and 3rd trimester, plus covariates) were divided into quintiles based on T. trichiura infection intensity in the second trimester. Mean hemoglobin concentrations and anemia prevalence in the third trimester were calculated for each group. Mean hemoglobin concentrations in the third trimester of each T. trichiura quintile were compared to the lowest quintile using generalized linear model (GLM) analysis. The prevalence of anemia, defined as hemoglobin ,11 g/dL [23], in the third trimester in each quintile was compared to that of the lowest quintile by logistic regression. Covariates found to be statistically significantly associated with the outcome were included in regression models: the model predicting hemoglobin levels included hookworm intensity and the model predicting anemia included hookworm intensity and the time interval between assessments for hemoglobin levels [22].

Results
Among the 935 pregnant women included in the analysis, 82% were infected with Trichuris trichiura, and 43% were co-infected with T. trichiura and hookworms. The highest T. trichiura infection intensity was 25,200 epg. Participants' characteristics are described in more detail elsewhere [22].
Women in the lowest three T. trichiura intensity quintiles had similar hemoglobin concentrations, with arithmetic mean levels of 11.53, 11.55 and 11.58 g/dL, respectively. In contrast, the fourth and fifth quintiles had significantly lower mean hemoglobin

Author Summary
Infection by the soil-transmitted helminth Trichuris trichiura is defined as 'light', 'moderate' and 'heavy' depending on its intensity. However, these intensity categories were developed in the 1980s, before any epidemiological data were available on the association between specific T. trichiura infection intensities and adverse health outcomes. Here, we re-analyzed data from a study of T. trichiura infection and anemia in pregnant women to determine the threshold (i.e. the lowest infection intensity) associated with an increased risk of anemia. Women with T. trichiura infections of intensities ranging from 601 to 1632 eggs per gram of feces (epg) (ie. a 'moderate' level of intensity) had a significantly higher prevalence of anemia and a significantly lower hemoglobin level than the reference group (i.e. women with T. trichiura infections of intensities ranging between 0 and 24 epg). This finding contrasts with the common belief that only 'heavy' T. trichiura infection (10,000 epg and above) can cause anemia. concentrations than the reference group (i.e. 11.24 and 11.05 g/ dL, respectively) ( Table 1). The fourth and fifth quintiles also had a statistically significantly higher risk of anemia, with adjusted odds ratios of 1.67 (95% CI 1.02, 2.62) and 1.73 (95% CI 1.09, 2.74), respectively ( Table 2).

Discussion
The fact that a statistically significant association between T. trichiura infection and anemia was found in this study, but not in any other study of pregnant women, can be explained, in part, by the fact that this time the association between T. trichiura and anemia was determined in a population of women who had received daily iron supplements. Therefore, the fraction of anemia attributable to an insufficient dietary intake may have been reduced in the study population, resulting in an increased fraction attributable to T. trichiura. This likely strengthened the association between T. trichiura and anemia in our study population, a finding that may not have been easily observable in other populations.
The 601-1632 epg T. trichiura infection intensity category was the lowest epg category where a statistically significant association between hemoglobin and anemia was found. This indicates that the threshold for the T. trichiura effect on hemoglobin and the risk of anemia in iron-supplemented pregnant women appears to be somewhere between 601 and 1632 epg. In other words, ironsupplemented pregnant women with ''light'' or ''moderate'' T. trichiura infection intensities, based on the current classifications, may indeed be at an increased risk of morbidity from anemia as a result of the infection. This finding has implications for STH control programs, in particular, those programs targeting pregnant women, because the efficacy of the commonly used deworming regimens of single-dose albendazole or mebendazole against T. trichiura is not optimal [25].
This analysis also provides support for categorizing a T. trichiura infection $1,000 epg as ''moderate'', as currently defined by WHO. In addition, for pregnant populations, even if they are receiving iron supplements during pregnancy, it may be that 601 epg should be considered a lower limit for this 'moderate' category.
The most important implication of these analyses is that moderate T. trichiura infection in pregnant women is a significant risk factor for anemia which, in turn, increases the risk of adverse maternal and infant health outcomes. Therefore, in a pregnant population where there is a high prevalence of T. trichiura infection and where intensity levels exceed 600 epg, it may be that additional care options beyond the commonly used single-dose albendazole or mebendazole should be considered. Table 2. Association between Trichuris intensity in the second trimester and anemia risk in the third trimester.