Multiple overlapping risk factors for childhood wheeze among children in Benin

Asthma is a chronic lung disease characterized by reversible airway obstruction resulting from an allergic reaction or hypersensitivity resulting in breathing difficulty. The symptoms include wheeze, shortness of breath, chest tightness and cough that vary over time and in intensity [1]. All age groups are affected, but children bear the greatest burden [2]. Asthma is among the top 10 causes of chronic conditions in the global ranking of disability-adjusted life years in participants 5–14 years [3].

In sub-Saharan Africa, infectious diseases continue to be a major driver of overall disease burden in children, although the region is undergoing both an environmental and an epidemiological transition toward an increase in non-communicable diseases (NCD), including increases in allergic diseases and asthma [4‐6]. In 2016, air pollution was the second largest risk factor causing NCDs globally, just below tobacco smoking [7], and children aged 5–15 years were three times more likely to die from the combined effects of ambient and household air pollution in Africa’s low- and middle-income countries (LMIC) than globally (12.9 versus 4.1 per 100,000) [8]. Air pollution was responsible for 1.1 million deaths across Africa in 2019, representing 16.3% of all deaths on the continent. The link between air pollution and asthma is complex. Studies have shown that air pollution may induce or aggravate asthma [9].

Pollution from road traffic predominates in urban areas, but rural populations are not safe from exposure to pollution. In 2019, household air pollution from cooking fuels accounted for 697 000 deaths, and ambient air pollution for 394 000 [10]. In LMIC, nearly three billion people rely on biomass for cooking in open stoves inside homes with little or no ventilation [11]. The populations most at risk are women and the young [12]. Biomass combustion produces high levels of pollution with many toxic agents [13], including many fine particles (PM2.5, PM10, etc.) but also carbon monoxide, nitric oxide and volatile organic compounds [14] that are known to be an important risk factor for respiratory allergic diseases including asthma.

People in rural areas of sub-Saharan Africa live not only with outdoor and indoor pollution but also with endemic parasitic diseases. Among the latter, helminths play a prominent role due to their chronicity allied to their ability to regulate the human immune system [15]. Helminth infections are characterized by the induction of Th2-type immune responses resulting in increased interleukin (IL)-4, IL-5, IL-13 and immunoglobulin (Ig) E levels [16]. The same type of response develops in some individuals on contact with allergens, leading to strong inflammatory responses as seen in asthma [17]. However, to ensure their own survival, in their chronic phase helminths establish an immunoregulatory network that limits the Th2-type response of the host. Thus, regulatory T and B cells are activated and high levels of the regulatory cytokine IL-10 are produced. These immunoregulatory elements not only maintain the helminth infection in the host, but also suppress inflammatory responses in infected individuals [18].

It has been hypothesized that increased hygiene, reduced family size, and consequently decreased microbial exposure levels, could explain the increasing global prevalence of asthma [19]. According to this ‘hygiene hypothesis’ reduced exposure to infectious agents may explain the increased incidence of allergic and autoimmune diseases in industrialized countries [20, 21]. The evidence linking helminth infections with allergic diseases is controversial. While several studies have shown inverse associations between helminths and allergy [22‐25], a recent review suggested that Ascaris lumbricoides infections may increase the risk of bronchial hyperreactivity in participants and of atopy in adults [26]. Notably, helminth elimination as a public health problem is integrated into the WHO roadmap for Neglected Tropical Diseases 2021–2030 [27].

To evaluate the potential contribution of different factors to wheeze in children, we took advantage of the DeWorm3 trial. DeWorm3 is a multi-country (Benin, Malawi and India) community-based cluster-randomized trial which aims to assess the feasibility of interrupting transmission of soil-transmitted helminths (STH) using community mass drug administration with albendazole, compared to standard of care school deworming [28]. The specific objectives of the study described here were, therefore: (i) to determine the proportion of participants with at least one episode of wheezing since birth and in the last 12 months among participants infected or not with STH, (ii) to characterize the exposure of participants to indoor air pollution and (iii) to assess the association between wheezing, exposure to indoor air pollution and STH infection status.

In the study described here, we investigated the factors associated with asthma symptoms in children living in a rural/semi-urban setting of southern Benin, focusing on soil-transmitted helminth (STH) infections as well as exposures to environmental factors, such as air pollution and allergens. Our study is thus distinctive—we know of no other published study to have assessed such a combination of factors in the context of asthma symptoms in a sub-Saharan African setting. The study leveraged the DeWorm3 trial in which the study subjects were participants. DeWorm3 is designed to assess STH transmission interruption through community mass deworming [28, 43].

In our study population the proportion with a history of wheeze was 5.2% and with current wheezing of 4.6%. Similar prevalences have previously been reported in children living in northern Benin [44]. In the Phase 3 ISAAC study, the prevalence of current wheezing in the African settings assessed was 10% and 14% in 6–7 year and 13–14 year age groups, respectively [31]. The Phase 3 ISAAC Centers assessed urban residents as opposed to the predominantly rural residents of our study site, a difference that, we speculate, possibly explains the difference in proportions of those found with wheezing. The prevalence of severe wheezing was 3.1% in our population, and the ratio between current wheezing and severe asthma is, therefore, consistent with the results of the third ISAAC study [31], showing that the proportion of those with current wheeze presenting clinical signs of severe asthma was higher in Africa than in other regions. The reasons could be poorer asthma care in low income countries together with lower awareness of wheeze being a symptom of asthma. However, differences of exposure to both air pollutants and infectious agents may also contribute to the greater severity observed in LMIC [31]. This difference underlines the importance of performing more studies in Benin and more generally in LMIC using the ISAAC questionnaire and tools.

The marked positive association we found between symptoms of wheeze and A. lumbricoides infection, with no association—either positive or negative—found for hookworm, are observations that are consistent with the findings of other studies conducted in sub-Saharan African settings [45, 46], although inverse associations between hookworm and allergic disease have nevertheless been reported in Uganda, for example, [47, 48]. Since a seminal study in Gabon on the topic [49], with the principal findings subsequently reproduced elsewhere [50], helminths in general have been thought to offer protection from allergic diseases. This is related to the fact that, to ensure their survival, these parasitic worms modulate the human immune system in the chronic phase of their infection through induction of IL-10-driven immunoregulatory responses that act, in a so-called ‘bystander’ way, to suppress allergic inflammation [15‐17]. Thus, chronic helminth infections may protect from allergic diseases including asthma [51‐54]. In such a scenario our results with respect to wheeze and A. lumbricoides could be seen as counter-intuitive, but they are, nevertheless, entirely in accordance with the conclusions drawn by reviews of recent studies on the topic of associations between STH and symptoms of asthma, including those of a systematic nature with meta-analyses [26, 55, 56].

Mechanistically, a precise immunological explanation for the association between wheeze and A. lumbricoides remains obscure, but a relationship with the inflammatory response to the pulmonary passage of larvae of A. lumbricoides seems the most plausible. It is known, for example, that Ascaris larval migration through airways may be associated with respiratory symptoms, such as wheezing, dyspnoea and bronchospasm [57, 58]. Separately, Ascaris is also associated with increased Th2-type and IgE responses to cross-reactive house dust mite(HDM)-specific allergens [59, 60], responses that themselves represent a positive risk factor for asthma and asthma severity [61‐63]. That being said, it should be borne in mind that our study population was participating in the DeWorm3 trial involving repeated treatment with the anthelmintic albendazole. Purely from the perspective of STH infections, then, any treatment-mediated decline in the prevalence of A. lumbricoides would be expected to be accompanied by a parallel decrease in the prevalence of wheezing. Although, as mentioned earlier, the prevalence of wheezing we observed here is similar to that reported in northern Benin, we do not know if it was actually higher in our study participants prior to implementation of anti-STH treatment in the study site. What remains also unclear is the speed with which such asthma-related symptoms would anyway be expected to resolve following clearance of worms. That wheezing was not associated with the number of treatments children received in our study over the 2 years prior to administering the ISAAC questionnaire could be interpreted as evidence that symptoms do not resolve rapidly. In the same context, intensive anti-STH treatment given every 3 months over a 2 year period in a rural area of Indonesia with a very high worm burden did not affect reported allergy symptoms [64], although it should be noted that the prevalence of STH infections remained relatively high in that study’s population.

We found a strong positive association—independent of STH infection—between wheeze and either exclusive use of open cookstoves, or use of palm cakes for lighting fires, findings that are consistent with earlier studies in Latin America [65, 66]. Open cookstoves use principally biomass as fuel, the incomplete combustion of which generates high levels of pollutants that are harmful to the lung [14]. From a mechanistic perspective, air pollutants cause oxidative stress to the airways, leading to inflammation, remodeling, and increased risk of sensitization [67]. The degree of openness of the cookstove is positively associated with the level of personal exposure to fine particulate matter [68]. Thus, improved cookstoves offer better combustion of the fuel by providing an insulated combustion chamber around and above the fire, enhancing the temperature of the fire and resulting in decreased CO and PM levels compared to open cookstoves [69‐71], associated with a reduction in wheezing symptoms both in mothers [72] and children [73]. In contrast to earlier studies [74], here we found no association between wheezing and indoor exposure to tobacco smoke, probably due to the comparatively low prevalence of tobacco smoking in our setting. We also found that being overweight was also positively and independently associated with current wheezing and severe wheezing episodes. Similar findings in children under five have been reported [75]. It is known that fat accumulates inside the airways of overweight individuals, altering their normal structure and leading to inflammation. There is also evidence of positive correlations between BMI, adipose tissue area on airway walls, airway wall thickness and the number of inflammatory cells [76].

There are a number of limitations of our study. One is that both exposure (air pollution, diet, allergens etc.) and wheeze were assessed based not on objective measures but on a questionnaire that could generate memory bias. The ISAAC questionnaire we used is nevertheless a highly standardized tool that has been used internationally for decades, facts that more than 500 peer reviewed publications attest to. Another limitation concerns the trial’s blinding context that prevented use of participants’ post-intervention helminth infection status. The true helminth prevalence at the time of administration of the ISAAC questionnaire would, logically, be expected to have been lower than the baseline level. In addition, the relatively low prevalence of the outcome measure, wheezing, may also have limited the study’s power to detect potentially meaningful associations, while the repeated treatment of children with albendazole may have modified some of the associations detected by decreasing the intensity of helminth infections.

To our knowledge, this study is the first to investigate risk factors for wheeze taking into account the potential role of both helminths and household air pollution (HAP) in the same model. HAP induces airway inflammation that elicits asthma symptoms, a process that our findings are consistent with. In the context of the epidemiological transition phenomenon ongoing in LMIC-like Benin, our work clearly highlights the importance, in future studies, of considering helminth infection characteristics (species, prevalence, intensity), combined with assessments of individuals’ exposure to HAP as well as their biometrics, to better determine potential risk factors for wheeze in childhood and adolescence.