Nutrition and Growth in Chronic Disease

Anat Guz-Mark
Raanan Shamir


Significant chronic diseases presenting during childhood and adolescence can predispose children to undernutrition, further complicating their medical condition with poor growth, reduced final adult height, and metabolic bone disease. Besides the poor nutritional intake and reduced appetite common in these condition, other factors may contribute to the risk of poor growth including intestinal malabsorption, hormonal alternations, delayed puberty, high energy requirements, and chronic inflammatory state. For this chapter, we selected and reviewed 10 leading articles published over the last year, evaluating different nutritional and growth aspects in five major chronic diseases: inflammatory bowel disease, celiac disease, chronic kidney disease, cystic fibrosis, and cerebral palsy. Although, these articles are not covering all the literature published during the last year, they provide a comprehensive overview of contemporary clinical research, allowing for a wide discussion of the various presentations and consequences of nutrition and growth alternations in pediatric chronic diseases.

Inflammatory Bowel Disease

Adult height in patients with childhood-onset inflammatory bowel disease: A nationwide population-based cohort study

Comments: Comments on this article as well as the following two articles (Rinawi et al. and Gupta et al.) are incorporated in the comments on the article by Ashton et al.


Prevalence and predictors of growth impairment and short stature in pediatric-onset inflammatory bowel disease

Comments: Comments on this article as well as the two articles (Mouratido et al. and Gupta et al.) are incorporated in the comments on the article by Ashton et al.


Continued statural growth in older adolescents and young adults with Crohn’s disease and ulcerative colitis beyond the time of expected growth plate closure

Comments: Comments on this article as well as the two articles (Mouratido et al. and Rinawi et al.) are incorporated in the comments on the article by Ashton et al.


Growth failure is rare in a contemporary cohort of paediatric inflammatory bowel disease patients

Comments: Growth failure occurs in 10–40% of children with inflammatory bowel disease (IBD) and is more prominent in children with Crohn’s disease than in ulcerative colitis (UC) [1, 2]. Growth impairment may precede the clinical signs of bowel disease in pediatric patients. The pathogenesis of growth impairment and failure is multifactorial and combines disease-related factors – mainly chronic inflammatory process as well as chronic malnutrition, and treatment-related factors including the use of corticosteroids [3]. Chronic malnutrition may occur in IBD due to decreased intake (reduced appetite, symptoms induced by food and medical restrictions), as well as secondary to intestinal malabsorption, increased nutritional needs and increased intestinal losses. As common for many chronic inflammatory diseases, the growth hormone (GH)-IGF1 axis is negatively affected by the inflammatory process and the elevation of plasma proinflammatory cytokines, causing GH resistance and impaired linear growth [4]. Patients with pediatric-onset IBD are at increased risk for growth failure at presentation and throughout treatment, and for reduced final adult height.

The first two studies reviewed in this chapter by Mouratidou et al. and by Rinawi et al., report reduced adult height in cohorts of pediatric-onset IBD. Mouratidou et al. investigated the Swedish National Patient Register and identified 4,201 patients diagnosed with pediatric-onset IBD between the years 1990–2014, with documented height measurement at adulthood (defined at the age of 18 years for females and 20 years for males). The study group included 59% males, 39% with the diagnosis of Crohn’s disease. The mean age for IBD diagnosis was 15 years in this cohort. The mean height at adulthood of the study group was 0.9 cm shorter than reference population, as well as 0.8 cm shorter than their healthy siblings. Growth retardation, defined as >8.5 cm of height below the target height, was identified in 5.0% of Crohn’s disease and 4.3% of UC patients, compared to 2.5% of matched reference individuals. The association with lower adult height was stronger in patients with Crohn’s disease, in patients with IBD onset before puberty, and in patients with a more severe disease course (indicated by bowel surgery or longer in-patient treatment).

In the study by Rinawi et al. reviewed in this chapter, data were collected regarding height at diagnosis and at adulthood in 416 patients with pediatric-onset IBD (compared to matched-controls), treated in a tertiary center in Israel between the years 1981–2013. In this study, significant differences in final height were observed only
among males: males with Crohn’s disease and with UC were 1.9 and 1.5 cm shorter (respectively) compared to controls. Still, the majority (>80%) of patients reached
their target height (calculated from the mid-parental height adjusted for gender). Identified risk factors for short stature when reaching adulthood included male gender and IBD diagnosis before puberty. Both of these studies demonstrate mild, yet statistically significant, decrease in final height among population diagnosed with IBD in childhood.

The third study reviewed therewith by Gupta et al. examined height gain among patients with IBD beyond the time of expected growth plate closure, using the ImproveCareNow registry in the USA. Cessation of statural growth occurs with radiographic closure of the growth plates, defined as bone age (BA) of 15 years in females and 17 in males [5], and hence these chronological ages (CAs) were chosen in the study as references for “expected” cessation of growth. Growth beyond these CAs means a delay in the closure of growth plates (delayed BA) and continued growth potential. The study included 3,007 patients with IBD (75.8% with Crohn’s disease, 56.2% females), with median age at diagnosis of 13.3 and 13.7 years for Crohn’s disease and UC, respectively. Continued statural growth was observed in 81% of patients with Crohn’s disease and in 75% of patients with UC. Median CA at achievement of final adult height was 18.2 and 18.1 years in males with Crohn’s disease and UC, respectively, as well as 16.4 and 16.3 years in females with Crohn’s disease and UC, respectively. Height velocity curves were shifted to the right in patients with IBD compared to reference population. This study demonstrates that patients with pediatric-onset IBD continue their statural growth beyond the time of expected growth plate closure, reflecting delayed skeletal maturation. On the other hand, it also highlights the capacity for continued catch-up growth in late adolescence and early adulthood among these patients. Performing studies for skeletal maturation could be essential in patients with IBD who have not reached their target height, as BA is an important tool for proper interpretation of statural growth.

Finally, the fourth study reviewed herewith, by Ashton et al., focused on assessment of growth among children with IBD, at the time of diagnosis and over 5 years of follow- up, in a contemporary cohort from the UK. The study included 490 patients with IBD diagnosed from 2011 to 2018, with mean age at diagnosis of 12.87 years for Crohn’s disease, 13.02 years for UC and 12.79 years for IBD-unclassified. Mean height SDS was not reduced at any time point for any disease subtype, and there were no significant changes in mean height-SDS between diagnosis and 1, 2, or 5 years of follow-up. Despite the near-normal mean height-SDS, the percentage of Crohn’s disease patients with stunting (height-SDS below –2.0) was above expected – 6.4%, with a decrease during follow-up to 3.7% at 1 year, 1.4% at 2 years, and 1.9% at 5 years. Mean weight-SDS was reduced at diagnosis, –0.39 for the whole cohort and –0.65 for patients with Crohn’s disease. The mean weight-SDS approached 0 by 1 year of follow-up. Identified factors associated with poor growth included male gender, anti-TNF treatment, and surgical resection.

Overall, in this contemporary cohort of pediatric IBD, impaired linear growth was rare at presentation and along follow-up. Some weight deficits at diagnosis showed an average recovery by 1 year. Risk factors for impaired linear growth were similar to those in the previous studies. This cohort presents more favorable outcome of growth in pediatric IBD, which could imply an improvement in the care of these patients over the last decade, particularly due to the positive effect of biologic treatments on linear growth in pediatric IBD [6, 7].


Celiac Disease

Growth rate of coeliac children is compromised before the onset of the disease

Comments: Comments on this article are incorporated in the comments on the article by Soliman et al.


Linear growth of children with celiac disease after the first two years on gluten-free diet: A controlled study

Comments: Celiac disease (CD), the most common immune-mediated enteropathy, affects the integrity of bowel mucosa and causes malabsorption of macro- and micronutrients, as well as chronic intestinal inflammation. Symptoms of malabsorption represent the classical presentation of CD, but extraintestinal manifestations also occur, including delayed growth and delayed puberty. The pathogenesis of altered growth and short stature in CD is multifactorial and involves both nutritional deficits secondary to malabsorption, as well as hormonal factors related to the growth hormone/insulin-like growth factor (IGF)-I axis and possibly autoimmune involvement of the pituitary gland [8]. With the treatment of gluten-free diet (GFD), patients with CD can experience an increase in the levels of growth hormone, IGF-I, IGF-II, and IGF-binding protein-3, and a decrease in proinflammatory cytokines [9]. Most children with CD exhibit catch-up growth during the first few years of GFD [10]. Nonetheless, some patients with CD may fail to demonstrate catch-up growth despite good adherence to GFD and normalization of serological markers [9, 11], further emphasizing the complexity of the mechanisms related to poor growth that is caused by CD.

Several previous studies have suggested that altered growth patterns are recognized in children with CD even before the onset of the disease [12]. Interestingly, some candidate genes were found to be expressed at least 9 months before the appearance of clinical and serological signs in children later diagnosed with CD [13]. The first study reviewed in this chapter, by Auricchio et al., reports growth rates of children with CD before the onset of the disease, as part of the results of the PreventCD study. The PreventCD cohort monitored 944 children who had a first-degree relative affected by CD (and were HLA DQ-positive) from birth to 6 years of age, including
periodic anthropometric measurements, assessments of symptoms of CD, and periodic anti-transglutaminase antibody sampling [14]. CD was diagnosed in 113 (11.9 %) children by the age of 6 years, with the median age at diagnosis of 40 months (none of them in the first year of life). No differences were observed between those who did or did not develop CD in birth weight, length, or gestational age. Differences in growth patterns were noted between the groups during the first year of life, before
any symptoms or antibodies developed. Children who later developed CD had a significantly lower growth rate in weight and length/height z-scores, compared to those who did not develop CD, although none had significant growth failure at that time point. Some differences in weight and length SDS were noted at 4 months of age, before gluten introduction, but did not reach statistical significance.

This longitudinal prospective cohort further demonstrates that growth rates can be altered in children with CD before any clinical or serological markers of the disease
appear. It is plausible to assume in these cases that some degree of intestinal inflammation and disturbances in the hormonal axis may precede the appearance of CD antibodies and the diagnosis of CD. Although the differences in growth patterns before gluten introduction were not statistically significant in this study, the role of genetic and hormonal factors, other than gastrointestinal, could not be excluded in the complex pathophysiology of altered growth in children with CD. The second study reviewed in this chapter, by Soliman et al., examined the linear growth of prepubertal children with CD after 2 years on a GFD. Anthropometric and laboratory data of 30 patients on GFD, with median age of 7.4 ± 2.6 years, were collected over 1 year of follow-up. Anthropometrics were compared to healthy age matched children. At study baseline, the height (Ht)-SDS was –0.40 ± 1.2 in the study group, with 4 out of 30 children defined as stunted (Ht-SDS ≤–2), after 2 years of GFD. At the end of 1-year follow-up, 2 children remained stunted. The average change in Ht-SDS was 0.15 ± 0.4. During the follow-up period, the weight gain per day was on average or above, for age and sex, in 27/30 children. Both Ht-SDS and BMI-SDS increased significantly in the CD group versus controls. In the CD group, the change in Ht-SDS was correlated with the change in BMI.

This study demonstrates that a significant portion of children with CD continue to present a slow catch-up growth (by crossing up height centiles) even after more than 2 years on a GFD. The compliance with GFD in the study group was not reported in the publication, although it was mentioned to be periodically assessed. The authors did stress that “nutritional rehabilitation” was performed, which included counseling regarding GFD and required energy and protein intake to allow for catch-up growth, and a free supply of high energy and protein supplementation was given to all patients who had BMI-SDS ≤–1. This could probably be accounted for the good results reported in this cohort, emphasizing the importance of continued nutritional assessment and counseling in order to reach catch-up growth in children with CD even beyond the first few years of GFD initiation.


Chronic Kidney Disease

Determinants of statural growth in European children with chronic kidney disease: Findings from the Cardiovascular Comorbidity in Children with Chronic Kidney Disease (4C) Study

Comments: Chronic kidney disease (CKD) in children is highly associated with significant growth impairment. Moreover, growth failure is a sensitive indicator of adverse clinical outcome and patient mortality in CKD [15, 16]. Many factors contribute to growth failure in these patients, including abnormalities of the GH-IGF1 axis, renal osteodystrophy, chronic metabolic acidosis, hyperparathyroidism, poor nutritional state, and inflammation [17, 18]. In addition, other significant contributing factors can be attributed to congenital syndromes associated with pediatric CKD, birth parameters, as well as parental height [19]. In this current study by Behnisch et al. and in the 4C Study Consortium, the growth performance of children and adolescents with CKD from 12 European countries was assessed over 6 years of follow-up. The study population included 594 children with CKD, 130 of whom required dialysis. The mean (SD) age of participants was 11.7 (3.2) years, 68.5% males, with only 6.8% of the cohort diagnosed with a defined congenital syndrome.

At time of enrolment, the mean height-SDS was –1.57, with variable prevalence of growth failure between countries. Short stature was most prevalent in Turkey (45%)
and the UK (39%) and least common in Germany (7%), with malnutrition most common in Serbia (19%). The highest prevalence of overweight and obesity was observed in the UK (42%). While the type of underlying kidney disease showed no impact on height, patients with syndromic disorders were 0.3 SD shorter than those with isolated kidney disease. The height SDS was positively associated with parental height, pubertal status, and treatment with recombinant GH. GH treatment was administered to 15% of patients in this cohort. This multicenter study emphasizes that growth failure in children with CKD is still a common problem, despite improvements in the management of CKD and its complications. The country of residence was identified in this study as an independent predictor of height in this population, other than the additional factors of parental height and syndromic diseases. Although not directly related to the results described above, as the treatment with recombinant GH is proved to be efficient in pediatric CKD with altered growth, its usage worldwide should be better utilized.


Cystic Fibrosis

Bone health in children and youth with cystic fibrosis: A systematic review and meta-analysis of matched cohort studies

Comments: Comments on this article are incorporated in the comments on the article by Gur et al.


Understanding the interplay between factors that influence bone mineral density in CF

Comments: Cystic fibrosis (CF) is an autosomal recessive monogenetic disorder, with a wide spectrum of phenotypic presentation characterized by progressive lung disease, exocrine pancreatic insufficiency, gastrointestinal abnormalities and malabsorption, impaired growth, sinusitis, and diabetes [20]. CF is strongly associated with poor nutritional status linked to high energy needs, energy losses, decreased nutrient intake and absorption, and recurrent infections and inflammation [21]. Decreased bone mineral density (BMD) and increased risk of fractures are well documented in patients with CF, which can be explained by multiple factors including the poor nutritional status and nutrient malabsorption, physical inactivity, use of glucocorticoid therapy, chronic pulmonary inflammation, and clinical status [22]. The unique study by Ubago-Guisado et al. reviewed herewith aimed to evaluate the longitudinal changes in areal BMD (aBMD) accrual during childhood and adolescence, compared to the longitudinal changes in healthy children. A systematic review and meta-analysis were performed that included 6 studies of patients with CF and 26 studies of healthy participants (controls), between the ages of 4 and 18 years. The median length of studies was 24 months and 18 months, for CF and healthy control studies, respectively. Changes in aBMD from baseline to end of follow-up were compared separately for whole body (WB), lumbar spine (LS), and femoral neck (FN). The results of this study showed no significant differences in the longitudinal changes of aBMD between children and adolescents with CF and healthy controls. Among subjects with CF, the pooled effect sizes (ES) for the change in aBMD were 0.29, 0.13, and 0.09 in WB, LS, and FN, respectively, whereas among healthy controls the ES for the change in aBMD were 0.37, 0.13, and 0.52 in WB, LS, and FN, respectively. Age, body mass index, height, and weight were not related to the changes in aBMD for either children with CF or healthy children.

The meta-analysis lacks data regarding the different stages of skeletal maturation and some clinical data regarding the participants in both groups. Also, it was mentioned that patients with CF in the included studies were standardly treated with multivitamins and mineral supplementations and pancreatic enzyme supplements, which may reflect good nutritional management of the participants with CF and explain in part the lack of differences in aBMD between groups. Moreover, the number of included studies of patients with CF was relatively limited. Besides these limitations, this study suggests that longitudinal aBMD development might not differ significantly between children with CF receiving medical care, compared to their healthy peers in the pediatric age group. A plausible explanation is the better care of children with CF; however, lack of data on the differences in the care between patients and controls leaves this possibility to be better shown.

The second study reviewed in this chapter by Gur et al. aimed to evaluate the correlation between BMD and a wide range of parameters in patients with CF. The study included 40 patients with CF, with a mean age of 18.3 ± 8.1 years. In this cohort, osteopenia was identified in 37.5% and osteoporosis in 27.5% of patients. The additional evaluated parameters included disease clinical score of modified Shwachman-Kulczycki (SK), lung functions, 6-min walk test (6MWT), hand grip strength (HGS), dietary
intake, pancreatic status, daily physical activity, and quality of life (QOL) questionnaires. In this study, BMD correlated with HGS, fat-free mass index, and QOL scores. BMD was similar in pancreatic sufficient and insufficient patients. Daily physical activity, nutrition intake, and 6MWT did not correlate with BMD. Age and SK score were found to be the best predictors of spine BMD. This small single-center study demonstrated a significant portion of patients with CF who exhibit low BMD, irrespectively of their pancreatic status or nutritional intake. This may suggest that the severity of the clinical disease may play a larger role in the pathogenesis of metabolic bone disease in CF, rather than malabsorption. As clinical scores of disease severity correlated well with reduced BMD, they could be valuable in simple and early identification of patients requiring evaluation of bone status.


Cerebral Palsy

Nutritional red flags in children with cerebral palsy

Comments: Children with cerebral palsy (CP) are at particular risk of malnutrition and impaired growth, further complicating their chronic neurological disability. Besides growth failure, undernutrition in children with CP may alter their immune function leading to increased risk of infections, increased risk of pressure sores, diminished respiratory muscle strength, and decreased cerebral function further compromising the ability of rehabilitation and development [23]. Multiple factors predispose children with CP to undernutrition, including significant feeding difficulties secondary to oropharyngeal incoordination and to frequent vomiting, increased energy requirement due to spasticity and involuntary movements, and high frequency of recurrent infections and respiratory difficulties [24].

The European Society for Pediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) published specific guidelines for the management of the gastrointestinal and nutritional complications in children with neurological impairment [25]. In these guidelines, five warning signs of undernutrition in this specific population were pointed:
physical signs of undernutrition, weight for age (WFA) z-score <–2, triceps skinfold thickness (TSF) <10th centile, mid-upper arm fat or muscle area (AMA) <10th centile,
and faltering weight and/or failure to thrive. This current study by Huysentruyt et al. evaluated the nutritional status of children and adolescents with CP in Belgium using different anthropometric indicators (including WFA, TSF, and AMA). A total of 325 patients with CP (61.5% males, median age of 11.7 years) were included and followed for 1 year. The anthropometric measurements were performed at baseline (T1), 6 months (T2), and 12 months (T3). Participants’ Gross Motor Function Classification System (GMFCS) score was distributed as follows: 15.4% with GMFCS-1, 29.2% with GMFCS-2, 15.1% with GMFCS-3, 16.6% with GMFCS-4, and 23.7% with GMFCS-5. At T1, 42.1% had at least one nutritional red flag (defined according to aforementioned ESPGHAN’s guidelines), at T2 40.3%, and at T3 41.4%. The median WFA z-score was –1.13, and in 71.4% of children with GMFCS-5, it was below –2. All anthropometric indices tended to decrease with increasing GMFCS (p < 0.001). The presence of dysphagia was also associated with lower scores on nearly all nutritional indices. During follow-up, weight loss was observed in 14.7 and 18.8% of those with at least one nutritional red flag at T1 or T2, respectively. It is important to note that in this cohort, gastrostomy was used in 6.8% of the cohort and in 22.1% of patients who had the highest score of 5 using GMFCS. None of the children with a gastrostomy lost weight during the study period.

This large prospective study demonstrates the high prevalence of undernutrition and warning signs of poor nutritional state in a contemporary cohort of children and adolescents with CP. Even in the presence of such warning signs, one-fifth of these children lost further weight during follow-up. Although a small subgroup in this study, the use of gastrostomy tube can probably offer better management in children with significant feeding disorders in this population. Children with the most severe neurological impairment, with a GMFCS score of 5, are at particular risk for malnutrition and merit special attention in order to avoid further complications.