1. IntroductionInflammatory bowel diseases (IBD), including Crohn’s disease (CD), ulcerative colitis (UC) and inflammatory bowel disease-unclassified (IBD-U), are chronic immune-mediated, relapsing-remitting disorders of the gastrointestinal tract. Incidence of pediatric IBD is growing worldwide, including in Hungary [1]. Malnutrition is frequently observed in pediatric patients with IBD. About two-thirds of patients are undernourished at the time of diagnosis, and they have an increased risk for morbidity and mortality [2,3]. Poor nutritional status may result in linear growth retardation that is considered as an extraintestinal manifestation in pediatric IBD [4]. Prevalence of moderate and severe underweight among children and adolescents in the general population is as low in Hungary as in other Western and Central European countries [5]. A recent WHO-initiated survey of 6-to-9-year-old children found 3.8% of Hungarian boys and 2.3% of Hungarian girls thin [6]. The prevalence of Hungarian children between the age of 1 and 5 years with body weight below 3rd centile varied between 1.70% and 3.16% in the age groups 1, 3, and 5 years of age, respectively, and in three consecutive years from 2013 to 2015 [7]. The relationship between undernutrition and IBD is well known. Severe undernutrition is one of the extraintestinal manifestations of Crohn’s disease, so we focused on the less well characterized associations of obesity and IBD.The other side of the malnutrition spectrum, obesity, is an existing socio-epidemiologic problem in all ages, particularly in the Western world. Growing prevalence of childhood obesity is a major public health concern of the century. In 2016, 124 million children between the ages of 5 and 19 years were considered to be obese worldwide [5]. The prevalence of childhood obesity in the USA is 18.5% [8]. In Europe, the prevalence of obesity ranges from 10% in Northern Europe to more than 40% in Southern Europe, affecting 2.1–19.8% of girls and 1.8–19.9% of boys below the age of 10 years in different countries [9]. The rate of obesity among adolescents has also increased between 2010 and 2018 in European countries, mainly in Southern Europe [10]. In Hungary, a population-based survey found that 6.6% of children and adolescents were obese [11], Factors associated with obesity among children and adolescents have not been extensively studied in Hungary. Similar to Cyprus and Italy, and in contrast to five other European countries, no association was found between parental socio-economic status and overweight of children in Hungary according to results of the IDEFIC consortium [12].The role of obesity in the etiology of IBD is not fully understood, especially in the pediatric population. Several clinical observations found an association between obesity and IBD in adults. Young women were suggested to have an increased risk for developing CD. Unfavorable clinical course of the disease, such as need for biologic treatment, loss of response to infliximab, perianal manifestation, earlier time to first surgery, and decreased quality of life were noted [13]. In contrast, the EPIC study, investigating males and elderly patients in addition to young females, did not find any association of obesity with disease location and phenotypic disease expression [14].Data on potential associations between obesity and pediatric IBD are scanty. In contrast to the results of adult cohorts, patients with UC were more commonly obese compared to patients with CD. North American cohorts with newly diagnosed IBD found 9–10% of pediatric patients with CD and 20–34% of patients with UC overweight [15]. In a Polish retrospective study, 1.9% of patients with CD and 8.4% with UC were obese at diagnosis [16]. A study from Israel found that pediatric patients with IBD both in the lower and higher ranges of the body mass index (BMI) spectrum had higher disease activity, more frequent disease exacerbations and more frequent need for anti-TNF-α therapy [17].Nutritional condition in the pediatric population has been characterized most frequently by BMI [16,17,18]. The recent guidance of the European Society of Pediatric Gastroenterology, Hepatology and Nutrition (ESPHGAN) also suggested to use weight, height, and BMI Z-scores for the evaluation of the nutritional status in pediatric patients with IBD [19]. However, cut-off values defining obesity differ between individual trials and publications. Thus, studies may not be comparable. For example, in some studies, BMI was considered abnormal above +2 Z-score values, whereas in some others, BMI was considered elevated above the 75th, the 85th, or the 95th centile values [17,18,20]. Since Hungarian reference curves do not list 85th and 95th centile values, a comparison of nutritional status and risk assignment of Hungarian pediatric patients with IBD based on these cut-off values can be done only by approximation.Body weight (BW), weight-for-height (WFH) percentiles and Z-scores, and ideal body weight percent (IBW%) are less frequently used anthropometric tools among children with IBD to characterize nutritional status. However, in addition to BMI, some of these indices were successfully used to characterize nutritional status of children with cancer, cystic fibrosis, and chronic liver disease [21,22,23]. IBW has an important role in the accurate definition of chemotherapeutic drug dosages among obese children [24]. IBW% was found more sensitive than BMI to characterize nutritional status in children with cancer [21].

Prior to the present survey, no data on the prevalence of obesity in Hungarian children and adolescents with IBD were published. Moreover, no study has been aimed at investigating a potential association between obesity and disease activity of pediatric IBD so far. The findings of this research may open a new path for further analysis of possible interactions between obesity and IBD in pediatric patients.

Therefore, the aims of the present study were (i) to assess the prevalence of malnutrition among children with newly diagnosed IBD in Hungary with a special emphasis on the rate of obesity; (ii) to implement a complete set of anthropometric parameters, such as BW, WFH Z-scores, and IBW % in addition to BMI; (iii) to compare patients by disease type (CD vs. UC) based on their nutritional status; and (iv) to analyze possible associations between nutritional status of patients as determined by anthropometric parameters and disease activity at the time of diagnosis.

4. Discussion

The present cross-sectional study is the first one in Hungary that assessed the nutritional status of newly diagnosed pediatric patients with IBD focusing on the prevalence of obesity and on the relationship between nutritional status and disease activity. The data from this epidemiologic study were collected from HUPIR, a nationwide, prospective registry of pediatric patients with IBD. The main result of this investigation is that prevalence of obesity among IBD patients is similar to that of the general Hungarian pediatric population, and obesity has a bimodal association with disease activity in patients with UC.

Prevalence of both obesity and IBD are growing worldwide among adults and children [29,33]. As reviewed by Singh et al., there is an association between premorbid obesity and the risk of developing CD, but not UC, among adults [34]. A similar trend has also been observed in the pediatric IBD population, with the difference that the prevalence of obesity was higher among children and adolescents with UC rather than CD [15].Analyzing patient data of HUPIR in our present study indicated that undernutrition was less frequent among children both with CD and UC than found by other investigators from different countries [2,3]. The prevalence of obesity classified by BMI was in the same range among patients with IBD (2.28% in CD and 5.48% in UC) as among the general Hungarian childhood population (5.50% among boys and 7.50% among girls) [11].In Croatia, Sila et al. did not find any obese individuals among pediatric patients either with CD or with UC. The prevalence of obesity among healthy children was 8.20% [35]. In contrast, in a Polish multicenter, retrospective study, 1.90% of pediatric patients with CD and 8.47% with UC were obese as defined by centile values (≥95th centile). In parallel with this observation, 18.60% of healthy pediatric males and 14.50% of females were obese in Poland [16]. In recent studies from North America, the prevalence of obese pediatric patients with CD is about four times, and that of patients with UC is three times higher than in Hungary [20,36]. Differences between the North American and the Hungarian pediatric populations could be attributed to the fact of differences in dietary habits as these countries are geographically, traditionally, and socio-economically far away from each other. It seems, the prevalence of obesity in pediatric patients with IBD varies across countries even within the same geographic region, such as Hungarian, Croatian, and Polish data characterizing Central Europe and even more between countries in different continents, as pointed out by comparing Hungarian and US data.The nutritional status of patients was defined by anthropometric parameters according the most resent ESPHGAN position paper [19]. In our study most of the children had BM, WFH, BMI, Z-scores, and IBW% values within the normal ranges. Prevalence of undernutrition was higher than obesity among patients with CD. Undernutrition is a well-known extraintestinal manifestation of pediatric CD since the intestinal pathology and the inflammatory response associated with CD result in weight loss. This explains that the activity of the disease at the time of diagnosis was more pronounced among undernourished children. Obesity, on the other hand, was more frequent among patients with UC than with CD. Median disease AIs of undernourished and obese patients with UC were similarly high in contrast to well-nourished patients with UC. Polynomial regression analysis indicated—for the first time—an association between obesity as determined by anthropometrical indices and disease activity in pediatric UC, suggesting that obesity may have a negative effect in this disease. The association was particularly prominent when nutritional status was determined by IBW%.Our results support the concept that obesity, as a proinflammatory condition by itself, may contribute to the aggravation of disease activity in UC. Moreover, the fact of being obese may represent a risk factor for developing UC [15]. Hidden, intra-abdominal accumulation of fatty tissue is not typical in UC [37]. In contrast, the presence of mesenterial creeping fat is a known phenomenon in patients with CD, even in the well-nourished and undernourished subpopulation [38]. Therefore, under-, and well-nourished patients may also suffer from creeping fat-induced inflammation [39]. The fact that obesity is a proinflammatory condition may be of special importance in countries such as Hungary, where the ratio of obesity in the general pediatric population is considerably high. Obesity represents the accumulation of abnormal or excessive fat tissue, resulting in weight gain. In CD, where normal weight or undernutrition is more common than obesity, fat accumulation does not necessarily coincide with weight gain. Elevated expression of adiponectin was found in creeping-fat of patients with active CD [40]. Secretory adipose tissue changes in CD may be observed most frequently in the mesenteric adipose tissue (creeping fat) which, in addition to local changes, may exert systemic effects as it produces soluble inflammatory mediators, such as C-reactive protein (CRP) [41]. An elevation in CRP level increases PCDAI, which is the indicator of disease activity. In contrast, PUCAI does not include data on inflammatory biomarkers, suggesting that inflammation is not a factor of disease activity in pediatric UC. Although evaluation of CRP levels was not included in this study—because of the lack sufficient data—the observed association between obesity and high PUCAI score at diagnosis may suggest a so far unrecognized relationship between obesity, inflammation, and pediatric UC.A clear correlation between disease activity and anthropometric parameters was found with using IBW% as a tool to determine nutritional status, not with the more generally accepted BMI, which is the recommended and most widely used index to define obesity among the general population and in pediatric patients with IBD [19,42,43,44]. The use of anthropometric parameters beyond BMI has been less commonly used but may be strongly recommended for some patient groups [21,22,23]. In this study, we assessed nutritional status based on BW, WFH Z-scores, and IBW% in addition to BMI Z-scores and found that the acuteness of these parameters was similar, except IBW%. For the calculation of IBW%, we used the 50th centile WFH value of healthy children. In the investigated population, more children proved to be obese when determined by IBW% compared to BW, WFH, and BMI Z-scores. IBW% identified 7.02% of patients with CD as obese, compared to 2.71%, 2.28%, and 2.69% assessed by BW, BMI, and WFH Z- scores, respectively. Among patients with UC, the rates of obesity were 12.17% by IBW% and 6.42%, 5.48%, and 4.80% by BW, BMI, and WFH Z-scores, respectively. This may be clinically relevant because defining obesity by IBW% does not depend on arbitrary cut-off values, in contrast to BMI.Detecting obesity as early as possible may be beneficial in influencing long-term outcomes in pediatric IBD. Determining IBW% in addition to the recommended BMI Z-score at regular patient visits may be helpful in governing dietary treatment and avoiding complications [19]. Recent studies found that not only undernutrition but also obesity, as determined by BMI, increased the risk of hospitalization and disease exacerbation in childhood IBD [43]. Von Graffenried et al. found that overweight and obesity were more prevalent among children with UC and were more frequently associated with perianal abscesses and surgery for this purpose among children with CD [45]. Although we have not studied disease course, including outcome of biological treatment if indicated, especially in patients with UC, or surgery in patients with CD, determining IBW% in addition to BMI may be a useful tool to predict complications in pediatric patients in IBD.BMI or IBW% determination alone, however, may not be sufficient to assess nutritional status. Examination of body fat distribution e.g., by whole body impedance analysis (BIA), computer tomography (CT), or magnetic resonance imaging (MRI) might have an important added value, as the visceral adipose tissue of non-obese patients with low BMI may also function as proinflammatory tissues [46]. Moreover, studying gut microbiota in patients with IBD may also have an important added value since a possible exciting relationship was revealed recently between obesity and changes is gut microbiota in patients with CD and UC [47]. The link between IBD and Western-style diet and changes in the gut microbiota was recently published [48].

The present study has a few limitations. One of the limitations is the lack of analyzing body composition with BIA, CT, or MRI. Anthropometric measures for evaluating subcutaneous adipose tissue accumulation, such as mid-upper arm circumference and skinfold thickness, were not registered in HUPIR. A further limitation of our study is that we did not investigate the effect of dietary habits of patients on nutritional status and disease activity. However, the major limitation is the cross-sectional nature of this study. Because of the lack of patient follow-up, we cannot define properly possible harmful effects of obesity at diagnosis of pediatric patients with IBD. Further prospective studies should be designed to assess the role of obesity in de novo pediatric IBD on disease outcome markers. Moreover, biomarkers of inflammation and of body composition should be involved in further investigations.