Projet de thèse en Sciences du vivant
Sous la direction de Jean-françois Huneau.
Thèses en préparation à Paris Saclay en cotutelle avec Gent University , dans le cadre de École doctorale Agriculture, Alimentation, Biologie, Environnement, Santé (2015-.... ; Paris) , en partenariat avec PNCA - Physiologie de la Nutrition et du Comportement Alimentaire (laboratoire) et de AgroParisTech (France) (établissement de préparation de la thèse) depuis le 12-05-2015 .
Résumé Contexte En 2014, 50 millions d'enfants de moins de 5 ans ont souffert de malnutrition aiguë, parmi lesquels 16 millions ont souffert de malnutrition aiguë sévère (MAS), la plupart d'entre eux vivant en Afrique Sub-Saharienne et en Asie du Sud-Est (UNICEF 2015). Les enfants atteints de MAS présentent un plus grand risque de mortalité (risque relatif entre 5 et 20). C'est l'un des facteurs sous-jacents de plus de 50 % des 10 à 11 millions de morts évitables par an chez les enfants de moins de 5 ans (OMS 2007, 2009). Chez les enfants âgés de 6 à 59 mois, les experts de l'OMS recommandent d'utiliser de manière indépendante deux critères anthropométriques pour le diagnostic de la MAS de type marasme (encore appelée émaciation sévère): un indice poids-pour-taille (P-T) < -3 (déviation exprimée en z-scores, ou écarts-type, par rapport à la moyenne de la référence de croissance OMS) ou un périmètre brachial (PB) < 115 mm (OMS 2009, 2013). Or il existe une divergence de diagnostic parfois très importante entre ces deux critères : ils n'identifient pas les mêmes enfants et ne renvoient pas la même image de la situation nutritionnelle dans la population, avec généralement un plus faible nombre d'enfants identifiés par le PB que par l'indice P-T. En dépit de la reconnaissance initiale par l'OMS de l'importance de l'hétérogénéité du diagnostic effectué par les deux indicateurs anthropométriques, et de la nécessité de poursuivre les recherches pour en comprendre les implications cliniques (OMS 2009), très peu d'études ont été menées sur cette question à ce jour (EN-Net 2013). En l'absence de critère de référence (« gold standard ») pour le diagnostic de la MAS, et d'investigations précises des besoins nutritionnels et des risques associés aux différents critères anthropométriques, il est difficile d'interpréter cette divergence de diagnostic. Or l'hétérogénéité de diagnostic pourrait logiquement correspondre à une hétérogénéité de besoins. Malgré ces incertitudes, il existe aujourd'hui une tendance forte à privilégier l'utilisation du PB pour le dépistage et l'admission des enfants à la prise en charge médico-nutritionnelle ; certains acteurs plaidant même pour une restriction des services de prise en charge aux seuls enfants identifiés par ce critère. Objectif Décrire et comparer le statut nutritionnel, métabolique, les processus physiopathologiques, et les risques associés aux différents types de diagnostic anthropométrique de la MAS, ainsi que leur réponse à la prise en charge. Structure de l'étude Nous conduirons une étude observationnelle prospective chez les enfants MAS qui seront détectés et référés vers les structures de santé, dans le cadre du programme de Prise en charge à base Communautaire de la Malnutrition Aigüe (PCMA). Les besoins nutritionnels et les risques associés seront évalués, à l'admission et au cours de la prise en charge, en utilisant une série d'indicateurs: - indicateurs indirects du statut nutritionnel, métabolique et immunitaire, parmi lesquels plusieurs marqueurs biologiques dont l'association avec le risque de mortalité a été récemment mis en évidence; - caractéristiques cliniques; - réponse au traitement en termes de taux et vitesse de guérison, et de rechute. Ces informations supplémentaires, collectées sur un sous-échantillon de patients, sont : - le taux de leptine dans le sang, indicateur d'une bonne quantité et fonctionnalité du tissu adipeux, et prédicteur de mortalité chez les SAM. - les ratios d'isotopes stables de l'azote et du carbone dans les cheveux, indicateurs du métabolisme présent et passé des réserves lipidiques (dépôts adipeux) et protéiques (muscles) - un ensemble de paramètres biologiques urinaires, indicateurs d'infections urinaires et du métabolisme lipidique et protéique - les paramètres de bio-impédance corporelle, indicateurs de l'hydratation, de la fonctionnalité de la masse maigre, et possiblement associés au risque de mortalité Méthodes de collectes d'informations spécifiques à l'étude OptiDiag. Les méthodes utilisées pour cette collecte supplémentaire d'information seront non-invasives et indolores. Dosage de leptine : le taux de leptine dans le sang sera renseigné par un test rapide effectué sur une goutte du sang. Ratios d'isotopes stables du Carbone et de l'Azote : les abondances naturelles en isotopes stables seront analysées dans une mèche de cheveux. Une mèche de cheveux sera prélevée à l'admission et une autre sera prélevée deux mois après la sortie du programme de prise en charge, lors d'une visite de suivi du projet MANGO. Paramètres biologiques urinaires : un test rapide par bandelettes urinaires multi-indicateurs (laboratoire Roche ou Analyticon) sera effectué sur un échantillon d'urine propre, à l'admission, puis à deux semaines et 8 semaines après l'admission. Cela ne sera effectué qu'à la condition que les enfants puissent uriner sur demande pendant la consultation, ou bien que l'urine puisse être collectée par pression des changes chez les plus jeunes enfants. Mesure de Bio-impédancemétrie : Les paramètres de bio-impédance corporelle seront mesurés de manière rapide non-invasive et indolore, par l'utilisation du de l'analyseur Quadscan 4000 (Bodystat, UK). Ces mesures seront réalisées à l'admission, puis à deux semaines, et enfin 8 semaines après l'admission. Sélection des participants pour les collectes d'informations supplémentaires OptiDiag
Biomedical Investigations for Optimized Diagnosis and Monitoring of Severe Acute Malnutrition: Elucidating the Heterogenious Diagnosis of SAM by Current Anthropometric Criteria and Moving Beyond
In 2014, 50 million children under 5 suffered from acute malnutrition, of which 16 million suffered from SAM, most of them living in sub-Saharan Africa and Southeast Asia. SAM children have higher risk of mortality (relative risk between 5 and 20). It is an underlying factor in over 50% of the 10 11 million preventable deaths per year among children under five (UNICEF 2015). At present, 65 countries have implemented WHO recommendations for SAM treatment (both in-patient for complicated cases and outpatient for uncomplicated cases) but these programs have very low coverage, reaching only around 10 15 % of SAM children (UNICEF 2013). In 2009 the World Health Organization (WHO) and the United Nations Children's Fund (UNICEF) issued a joint statement in an effort to harmonize the application of anthropometric criteria for SAM diagnosis and monitoring in child aged 6 59 months; the statement presents recommended cut-offs, and summarizes the rational for the adoption, of the following two anthropometric criteria: 1. Weight-for-Height Z-Score (WHZ): WHO and UNICEF recommend the use of a cut-off for weight-for-height of below 3 standard deviations (SD) of the WHO standards to identify infants and children as having SAM. Additionally, analysis of existing data show that children with a WHZ < -3 have a highly elevated risk of death (WHO 2009). 2. Mid-Upper Arm Circumference (MUAC): WHO standards for the MUAC-for-age show that in a well-nourished population there are very few children aged 6 59 months with a MUAC less than 115 mm. Children with a MUAC less than 115 mm have a highly elevated risk of death compared to those who are above. Thus it is recommended to [use] the cut-off point [of] 115 mm to define SAM with MUAC (WHO 2009). Table 1 below reports the recommended diagnostic criteria for SAM in children as per the 2009 joint statement on the identification of SAM in infants and children. 1.1.2. DIAGNOSTIC DISCREPANCY According to WHO experts, WHZ and MUAC can be used independently to indicate severe acute malnutrition (WHO 2009). There is however a significant and sometimes huge discrepancy between these two criteria: they do not usually identify the same children as acutely malnourished; moreover when used as proxy indicator to assess a deteriorating nutritional situation at a population level, these criteria do not report the same level of global acute malnutrition in the same zone. It was reported that only about 40% of SAM cases identified by one indicator are also diagnosed as such by the other (WHO 2009). For example, among severely malnourished children hospitalized in rural Kenya, 65.1% (486/746) of the WHZ 3 cases also had a MUAC < 115 mm, whereas 56% (489/873) of the MUAC < 115 mm cases were also identified by WHZ < 3 (Berkley 2005). In that study, 42.9% (489/1140) of the SAM cases were identified by both indicators. The discrepancy between the two indicators can be even more extreme (Laillou 2014, Fernandez 2010, Dairo 2012). Fernandez et al. reported that among 34,937 children between the ages of 6 59 months from 39 nutritional surveys, 75% of the children with a WHZ < 3 were not identified by a MUAC < 115 mm (Fernandez 2010). In Cambodia, this proportion was above 90%, whereas 80% of MUAC < 115 mm were not detected by WHZ < 3 (Laillou 2014). Most of the time, caseloads defined by WHZ are much larger than by MUAC, but the contrary may happen as well, especially in the younger age groups. 1.1.3. PROGRAMMATIC CONFUSION Such discrepancy generates important programmatic challenges and confusion (ENN 2012). On the one hand, a strategy where the diagnosis can be based on either indicator, as recommended by some authors (WHO 2009, Laillou 2014) may unduly inflate the workload of nutritional rehabilitation programs, as the most appropriate management of children identified by one indicator and not by the other is uncertain. On the other hand, relying on only one of these indicators, e.g. using only MUAC < 115 mm in community-based programs, may under-detect true acute malnutrition cases and result in missed opportunities to treat a severe condition (Laillou 2014, Grellety 2015). In recent years, however, the use of MUAC alone for admission has been discussed (Ali 2013, Goossens 2012, Roberfroid 2013), and is increasingly applied in a variety of contexts. In particular, more and more national protocols for SAM management consider MUAC only management as programmatic option. The national guidelines in Bangladesh, for example consider only low MUAC as an admission criterion for uncomplicated SAM management, which by de facto excludes a vast majority of the SAM children, those who have WHZ < 3 and MUAC ≥ 115mm. Many benefits of using MUAC exist: MUAC is predictive of death, easy to use, acceptable, and favors community-based screening methods (Briend 2010, Myatt 2009). Yet, as these two anthropometric tools select different children for treatment, as outlined above, this complicates the programmatic paradigm shift from admitting children using MUAC < 115 mm and/or WHZ < -3 to a new model admitting children using MUAC < 115 mm only. Depending on context, up to 6379% of children currently recommended for therapeutic feeding with WHZ < 3 and/or MUAC < 115 mm would not be eligible if using MUAC < 115 mm alone for admission (WHO 2009, Bern 1995, Laillou 2014). 1.2. RATIONALE To inform decision making regarding the use of MUAC as a standalone admission criterion in nutrition programming, more information on the programmatic and clinical implications of using MUAC alone is urgently needed. Despite WHO clearly highlighting the importance of this anthropometric diagnostic heterogeneity, and requesting more investigation (WHO 2009), very little has been done so far. Preliminary reports demonstrate demographic and anthropometric differences among children identified by WHZ and MUAC: MUAC is more likely to identify children that are younger, female and with concomitant stunting. These data have been used to suggest a role for MUAC to identify children that are potentially more vulnerable or at a higher risk of death, supporting the transition to a MUAC-only based admission criterion (Isanaka 2015). Recent secondary analysis of the clinical profile and outcomes of SAM children admitted to an outpatient SAM program in Niger infirmed this hypothesis, by showing a similar vulnerability profile in SAM children presenting with a MUAC < 115 mm (with or without concomitant WHZ < 3) and in SAM children with a MUAC ≥ 115mm, i.e. only with WHZ < 3, who would not be considered for treatment in case of MUAC-only programming (Isanaka 2015). Furthermore, according to this study, and a similar one from a SAM management program in South-Sudan (Grellety 2015), the anthropometric category of SAM children displaying the highest vulnerability at admission and the worse treatment outcomes are those combining MUAC < 115mm and WHZ < 3. These results are in line with previous observations from an in-patient SAM management program in Kenya (Berkley 2005). Beyond the investigation of possible variations in mortality risks, all reviews of available evidence on this issue highlight the need for robust research (ENN 2012; Roberfroid 2013) to further investigate the physiological significance of the different anthropometric criteria and to better understand how the clinical status and nutritional needs of the children are addressed over the course of nutritional rehabilitation. A key issue is indeed that these two different indicators identify different populations of children, the reason for which is unknown due to the lack of a gold standard. Current hypotheses to explain the diagnosis discrepancy are that: - WHZ<-3 overestimates the diagnosis of acute malnutrition in populations with a slender morphology (i.e. with a low sitting-to-standing-height ratio; SSR) as observed in pastoralists (Myatt 2009). - MUAC at a fixed cut-off overestimate SAM in the younger children, in girls and in the stunted children, and on the contrary underestimates SAM in older, male, and non-stunted children (Roberfroid 2013). Young age, being a girl and stunting are indeed factors known to be associated with lower MUAC measurements and were already shown to be independently associated with MUAC diagnosis (WHO 2009, Berkley 2005, Shams 2012): lesser levels of acute nutritional deficits and wasting might thus be necessary to reach the 115mm cut-off in these children. These hypotheses have recently been supported by the analysis of the strength of the association between these factors and the diagnosis discrepancy in nutritional cross-sectional surveys (Roberfroid 2015). WHZ and MUAC criteria also may identify a separate kind of physiological deficit. It has been hypothesized that this might be related to differing impairments of fat and muscle mass stores, with MUAC reflecting preferentially fat mass for some authors (Chomto 2006) and muscle mass for others (Brambilla 2000). An analysis of body composition in a cohort of Ethiopian infants recently confirmed WHZ as a good marker of tissue masses independent of length, while MUAC appeared more as a composite index of poor growth indexing jointly tissue masses and length (Grijalva-Eternod 2015). Children identified by different criteria may thus require different treatment, one that is tailored to nutritional deficit. For instance, lower anthropometric response to treatment (lower MUAC gain and weight gain, longer treatment duration and higher proportion of non-responders) has already been observed in younger, stunted girls identified by MUAC (Roberfroid 2013). This might be linked to a suboptimal response in less severely wasted children, or might be due to a higher proportion of false positives in this sub-population, or be an indicator that the treatment is less effective or required in such children. Also, a recent meta-analysis of follow-up datasets evidenced a dramatic increase in mortality risk in children combining low WHZ and stunting (MUAC was not factored in) (McDonald 2013). Today, in the absence of a gold standard for SAM, it is difficult to interpret different and often divergent anthropometric diagnoses. Additionally, there is a vital need to better understand if and how far physiological recovery, beyond anthropometric growth (which might be transient or sub-optimal) is achieved under the current SAM management strategy. Moreover, this understanding should encompass the whole population of children affected by anthropometric deficit, beyond just those few complicated cases that reach the hospital for inpatient nutritional rehabilitation. It should also account for potential contextual variation in the link between anthropometry and nutritional status. In order to describe and compare nutritional needs and risks associated with the different types of diagnosis as they are present in the community, we propose to conduct prospective cohort studies of SAM children who will be detected and referred to treatment in the catchment areas of community-based acute malnutrition management programs (WHO 2007). Such programs combine both outpatient and inpatient nutritional rehabilitation, and an effective community outreach component. Nutritional needs and risks will be evaluated using a range of indicators: - proxy indicators of nutritional, metabolic and immune status, among which several biomarkers whose association with risk of death has been recently evidenced in SAM children; - clinical characteristics; and, response to treatment in terms of cure rates, recovery speed, relapse. The indicators necessary to do so must be easy to collect with low invasiveness and should provide reliable information regarding the severity of nutritional status. 1.2.1. ISOTOPIC EVALUATION OF HAIR Isotopic analysis of stable carbon and nitrogen in human hair can be investigated and measured throughout the course of nutritional deprivation to reconstruct the onset and duration of undernourishment (Neuberger 2013) as well as tracing the temporal evolution of nutritional status (Hatch 2006, Mekota 2006, DeLuca 2010, Petze 2009, Neuberger 2013). Several studies have revealed that factors like diet, disease and injury can influence nitrogen isotope ratios (δ15N) in human tissues. Specifically, δ15N values reflect the nitrogen balance of an organism in that during a catabolic state (tissue breakdown) δ15N values increase while during an anabolic state (tissue buildup) δ15N values decrease (D'Ortenzio 2015). In contrast, carbon isotope ratios (δ13C) are shown to decrease during catabolism and increase during anabolism. Thus, during starvation the body becomes enriched in 15N and depleted in 13C at the same time. Since keratin remains unchanged after synthesis, and the speed of hair growth is constant (around 2.5 mm per week), weekly information on protein-energy metabolism can be traced back along the hair follicle, thereby indicating not only the severity of the episode of wasting but also the metabolic effects of the nutritional rehabilitation (on both lipid and protein anabolism). Isotopic evaluation of stable carbon and nitrogen in hair will therefore be used to create a retrospective timeframe of nutritional status and trace the physiological recovery of children during SAM management. 1.2.2. LEPTIN AND ADIPONECTIN A recent study using non-targeted metabolomics analysis to characterize changes a broad array of hormones, cytokines, growth factors and metabolites during the treatment of SAM has revealed that a major biochemical predictive factor for mortality is low-level leptin (Bartz 2015). Low leptin and adiponectin levels reflect the adequacy of fat stores. Depletion of white adipose stores is postulated to limit the ability of a child to sustain energy production during the course of the illness and thereby increase the child's risk of death. Alternatively, hypoleptinemia may reduce viability affecting glucose and energy homeostasis or immune competence (Bartz 2015) Leptin and high molecular weight (HMW) adiponectin targeted analysis will therefore be used to create a metabolic profile of SAM patients at presentation and during nutritional rehabilitation, and may predict mortality prior to and during treatment. Dr. Michael Freemark and colleagues at DUMC are currently developing novel point-of-care micro-assays to characterize the hormonal status of leptin and adiponectin in SAM children from a single fingerstick that will be piloted. 1.2.3. MICRONUTRIENT AND IMMUNE RESPONSE BIOMARKERS Deficiencies of vitamin A and iron are among the most common micronutrient deficiencies related to childhood undernutrition and are both linked to compromised immune function (Müller 2005). Manifestations of isolated iron deficiency include anemia, fatigue, impaired cognitive development and reduced growth and physical strength (Fleming 2003, Levin 1993, Nemer 2001, Thomas 2002). Vitamin A deficiency contributes to anemia by immobilizing iron in the reticuloendothelial system, reducing hemopoiesis and increasing susceptibility to infections (Müller 2005); it is essential for the functioning of the immune system and its deficiency has been clearly shown to be associated with diarrhea and related mortality (Fleming 2003, Demba 1998). Vitamin A deficiency has been evidenced as frequent in SAM children (Sattar 2012). Vitamin A status, measured by the surrogate marker RBP, has been shown to be low in SAM children and to rise during nutritional rehabilitation (Sattar 2012). Mean serum vitamin A has been shown to decrease with increasing stunting (HAZ), wasting (WHZ) and underweight (WAZ) (Marasinghe 2015). Additionally, there are indications that the storage levels of iron in SAM children are increased not decreased, even in the presence of quite severe anemia (Golden 2009). However, there is a major lack of evidence on this point; we know that these parameters also need to be adjusted for inflammation biomarker, anemia and malaria, which was not done in the studies mentioned by Golden (Thurnham 2015, Ianotti 2013, Bresnahan 2015, Wessels 2014). Immune response biomarkers like C-reactive protein (CRP) are elevated in SAM children with severe bacterial infections. CRP is therefore a potentially valuable clinical tool for identifying bacterial infections, and recent research has shown that a rapid CRP could be useful in field settings to identify children most at risk for dying, with a relatively good negative predictive value (81% sensitivity, 85% specificity) (Page 2014). There is a need to evaluate the relationship between micronutrient status, immune response and anthropometric diagnosis of SAM children at and to examine the extent to which nutritional rehabilitation is effective in treating deficiencies of vitamin A and iron and to prevent deficiencies during catch up growth. An inexpensive and sensitive simple sandwich enzyme-linked immunosorbent assay (ELISA) technique was recently developed to measure indicators of vitamin A and iron deficiency (Erhardt 2004). Due to the low cost, high throughput, and comparability to traditional tests, this procedure has several advantages for assessing vitamin A and iron status on the field. Moreover, it can easily be combined with the measurement of immune response biomarkers like CRP and α1-acid glycoprotein (AGP). CRP, AGP, as well as biomarkers of iron (serum ferritin and serum transferrin receptor) and vitamin A status (serum retinol binding protein) can be assessed in a few drops of capillary blood (Erhardt 2004). CRP and AGP will be used to adjust for the effect of inflammation on the micronutrient status indicators. Inflammation is indeed known to elevate serum ferritin and depress retinol binding protein as part of the biological acute phase response to inflammation (Tomkins 2003, Thurnham 2003, Thurnham 2010). 1.2.4. URINE TESTS The presence of ketones in the urine, indicating lipid catabolism (fat tissue disintegration and rapid weight loss) was evidenced during fasting and SAM (Bartz 2014, Jackson 2006). Metabolic status for SAM children at the time of enrollment in CMAM has been characterized by ketonemia; yet, lipolysis decreases in response to nutritional rehabilitation suggested by total ketones (Bartz 2014). Moreover, biomarkers of urinary infections like urinary nitrites and urinary leucocyte esterase (LE) have also been shown to be associated with an increased mortality risk in SAM children (Thuo 2010). Positive dipstick urinalysis administered as a bedside screening test for either nitrates or LE is associated with a higher case fatality and was shown to be a strong predictor of mortality in children admitted with SAM (Thuo 2010). Non-sterile urine sample will be also carried out when possible, and these biological parameters will be measured using through the urinary multiple indicator strips (e.g. Roche laboratory, or Combi Screen of Analyticon). 1.2.5. BIOELECTRICAL IMPEDANCE (BI) It has been suggested that different anthropometric diagnoses identify children with different body composition, and associated nutritional needs (Laillou 2013, Chomtho 2006 Grijalva-Eternod 2015, Brambilla 2000). Restoration of body composition indicates successful management of SAM (Girma 2015). Bioelectral impedance (BI) is a safe, rapid and easy technique often used to assess body composition, predicting total body water (TBW) in non-edematous children. It has demonstrated utility for indexing acute changes in hydration in children with SAM during in-patient treatment. This technique could also potentially distinguish tissue versus hydration relates weight catch-up during or post treatment. Lastly, BI analysis may predict survival outcomes for children hospitalized for SAM (Girma 2014). BI parameters will therefore be used to describe body composition at admission and the restoration of body composition throughout nutritional recovery. .