OUP user menu

Growth failure and intestinal inflammation

A.B. Ballinger , C. Camacho‐Hübner , N.M. Croft
DOI: http://dx.doi.org/10.1093/qjmed/94.3.121 121-125 First published online: 1 March 2001


At least one third of patients with inflammatory bowel disease (IBD), namely Crohn's disease and ulcerative disease (UC), first develop symptoms during childhood and adolescence. In contrast to the adult population, in which the incidence of Crohn's disease has plateaued, the incidence of juvenile‐onset Crohn's disease continues to increase.1 The clinical spectrum and severity of disease varies considerably in young people, as it does in adults, but there are a number of special problems that face children and adolescents with IBD. In this review we will discuss the prevalence, aetiology, pathogenesis and management of growth failure in these patients.

Prevalence and natural history of growth failure

Growth failure frequently complicates the clinical course of children with IBD, more often in Crohn's disease than in UC.2,,3 Its reported frequency depends to some extent on its definition; a reduction in height velocity is the most accurate measurement of growth. A decrease in height velocity below the third centile has been reported in as many as 88% of patients before the diagnosis of Crohn's disease; in just over half of this group, growth failure was documented before the onset of symptoms attributable to Crohn's disease.4 During the subsequent clinical course, 20–40% of patients continue to have severe linear growth retardation, defined as height below the third centile.3,5,,6 While the majority of adults who had onset of IBD in childhood have heights between the 5th and the 97th centile, the distribution is skewed towards the lower centiles (Table 1).7 Final height below the fifth centile is reported in 7–30% of patients.7–,10

View this table:
Table 1

Distribution of adult height percentiles in patients with early onset IBD7

Height percentilePatients (%)

Aetiology and pathogenesis of growth failure

Growth failure associated with IBD has been attributed chiefly to undernutrition.11–,13 The nutritional deficiency appears to be related primarily to inadequate calorie intake, rather than to increased gut losses or to an increase in energy expenditure. Energy intake in growth‐retarded children with Crohn's disease has been documented at only 42–82% of expected values. Nutritional supplementation reversed the calorie deficit and also increased growth velocity.12,,13

There is increasing evidence that the inflammatory process per se may directly inhibit linear growth and play a major role in the aetiology of growth retardation.14–,19 Enteral nutrition not only improves weight gain but suppresses active intestinal inflammation, reduces mucosal cytokine production and induces disease remission.17,,18 Moreover, after institution of enteral nutrition, significant changes in serum growth factors and inflammatory indices have been observed before any changes in nutritional parameters.19 The relative contribution of reduced calorie intake and inflammation to linear growth delay has been difficult to determine from human studies. We have measured linear growth in prepubertal rats with trinitrobenzenesulphonic acid (TNBS)‐induced‐colitis, a widely used and well validated model of human Crohn's disease, and compared the results to both healthy free‐feeding controls and a pair‐fed group (healthy animals whose daily food intake is matched to the colitic group, thereby allowing us to separate the effects of undernutrition from inflammation on linear growth).14 In colitic rats, we found that about 60% of the final growth impairment could be attributed to undernutrition, inflammation accounting for the remaining growth deficit (Figure 1).

Steroids, which are used in the treatment of IBD, may also cause growth failure, although the precise mechanisms are not known. However, in most centres enteral nutrition is the first‐line treatment in children with active Crohn's disease. It is as effective as oral steroids in inducing a remission in small‐bowel Crohn's disease, and growth velocity following treatment has been shown to be greater in the groups undergoing the enteral diet.17,,20 However, for some patients who require systemic steroids to control disease activity, the anti‐inflammatory effects may outweigh any inhibitory effects on growth. It is not known whether topical steroids, which are frequently used in the treatment of distal UC, also have a negative effect on growth, although only a small proportion of children with IBD present with this disease distribution. In children with asthma, inhaled budesonide has been shown to be associated with a dose‐related suppression of linear growth.21

Figure 1.

The change in body length (mean±standard deviation) in healthy free‐feeding controls, colitic and pair‐fed groups. The pair‐fed group are healthy, but food intake is matched to the colitic group who have hypophagia as a result of inflammation. Thereby, the effects of undernutrition (occurring equally in both colitic and pair‐fed groups) on linear growth are separated from the effects of inflammation (occurring only in the colitic group). The relative contribution of undernutrition and inflammation to the growth deficit in the colitic group are indicated by the solid and broken arrows respectively. *p<0.02 vs pair‐fed; p<0.002 vs healthy free‐feeding controls.

Endocrine mediators of growth failure

The integrity of the growth hormone/insulin‐like growth factor‐1 (GH/IGF‐I) axis is essential for normal linear growth.22,,23 GH stimulates production of IGF‐I from the liver (the major source of circulating IGF‐I), which increases plasma IGF‐I levels and local production of IGF‐I in growth plate chondrocytes.24 Although in health the mode of action of IGF‐I is disputed, growth retardation associated with fasting is thought to be mediated by the systemic endocrine mechanism.25 Young patients with IBD have normal stimulated and spontaneous GH secretion but reduced plasma IGF‐I concentrations, indicating hepatic GH insensitivity.16,26,,27 Indeed, this endocrine profile exists in other chronic inflammatory diseases associated with growth retardation, suggesting that similar aetiological/pathogenic mechanisms may operate.28,,29 Rats with experimental colitis also have reduced plasma concentrations of IGF‐I and normal GH. Comparison with pair‐fed controls suggests that, similarly to linear growth, the reduction in IGF‐I is mediated by a combination of undernutrition and active inflammation.14 In this model, exogenous administration of IGF‐I increased plasma IGF‐I concentrations and produced a parallel increase in linear growth, suggesting that a reduction in circulating IGF‐I does indeed contribute to the growth deficit.14 It is not known if children with IBD and growth failure show a similar response to IGF‐I administration. Abnormalities of thyroid function, which may also cause stunted growth, seem unlikely on the basis of studies in adults with IBD in whom serum free thyroxine concentrations were normal.30

Cytokines and growth failure

De Benedetti et al. have suggested that IL‐6 may be part of the mechanism by which chronic inflammation inhibits growth.31 Transgenic mice which overexpress IL‐6 are growth‐retarded and have reduced plasma concentrations of IGF‐I, despite a normal food intake. In this study, treatment of control mice with IL‐6 also resulted in a decrease in IGF‐I concentrations, thus confirming the relationship between elevated concentrations of IL‐6 and suppression of IGF‐I. Plasma concentrations of GH are normal in IL‐6‐transgenic mice (as they are in animals and humans with intestinal inflammation) therefore demonstrating that the effect on IGF‐I levels is not mediated indirectly via an effect on GH production. Preliminary studies suggest a similar relationship between IL‐6 and IGF‐I in the TNBS‐colitis model, in which immunoneutralization of IL‐6 normalizes liver IGF‐I mRNA expression and increases both plasma concentrations of IGF‐I and linear growth.32 Although plasma concentrations of IL‐6 are increased in rats with TNBS‐colitis14 and patients with IBD,33 it is not known if the systemic concentrations achieved are sufficient to suppress IGF‐I. Alternatively, the hepatocyte IGF‐I response may be to cytokines draining in the mesenteric vein from the intestine.

Tumour necrosis factor‐α (TNF‐α), another macrophage‐derived cytokine, may also be an important mediator of growth failure. Excessive production of TNF‐α caused growth failure in TNF‐transgenic mice, and TNF‐α has direct inhibitory effects on growth plate chondrocytes.34,,35 In vitro, administration of TNF‐α and IL‐1β (another pro‐inflammatory cytokine) inhibited IGF‐I production from rat hepatocytes by reduction of GH receptor synthesis.36 Plasma concentrations of TNF‐α are increased in children with Crohn's disease,37 and its key role in disease pathogenesis is illustrated by the beneficial effect of anti‐TNF‐α antibodies on disease activity.38 The role of TNF‐α, if any, in linear growth retardation has not yet been investigated in either humans with IBD or experimental models of intestinal inflammation.

The final common pathway for peripheral mediators of bone growth is the epiphyseal growth plate. The increase in length of a bone is determined by a complex interplay of growth plate chondrocyte proliferation, matrix synthesis and degradation, and chondrocyte maturation and hypertrophy. In rats with experimental colitis, histological examination of growth plate morphology has shown reduced heights of the proliferative zone and terminal hypertrophic chondrocytes compared to those of pair‐fed controls.39 These observations are consistent with abnormalities of both chondrocyte proliferation and maturation, although no detailed studies have been performed. Furthermore, it is not known if suppression of growth plate function is mediated entirely by a reduction in circulating IGF‐I. In vitro studies suggest that cytokines may also have direct inhibitory effects on growth plate chondrocytes.35 The proposed mechanisms leading to linear growth failure are represented diagrammatically in Figure 2.

Figure 2.

Schematic diagram representing the proposed mechanisms of linear growth retardation associated with intestinal inflammation. Undernutrition and pro‐inflammatory cytokines exert an independent inhibitory effect and lead to hepatic GH resistance and suppression of circulating IGF‐I. They may also directly inhibit growth plate chondrocyte proliferation and maturation.

Management of growth failure

It is essential that height, weight, puberty staging and bone age are accurately and regularly measured and recorded in young patients with IBD. A recent study has shown under‐recording of these variables of growth in children with IBD.40 A fall in height velocity may be the first indication of disease relapse, and may be accompanied by only minor gastrointestinal symptoms.

From the preceding discussion it seems likely that optimal management of growth failure involves calorie supplements (to correct undernutrition) and anti‐inflammatory therapy to reduce cytokines which inhibit linear growth. Exclusive enteral feeding with elemental or polymeric feeds for 6 weeks has the major advantage of combining anti‐inflammatory properties with an increase in energy intake, and is thus ideal for patients with growth failure. Although steroids also induce clinical remission the height velocity remains lower than in enterally‐fed children, making them the second‐line option.17,,20 Polymeric diets, when compared with elemental and semi‐elemental diets, are as effective in inducing disease remission,41 are more palatable and thus generally do not need to be administered intragastrically. In patients with UC, enteral nutrition may be necessary to correct undernutrition, but it has no role in induction of disease remission.

About 40–50% of patients with Crohn's disease will relapse in the first year after induction of disease remission.42 Maintenance of remission, particularly during periods of rapid growth, e.g. during puberty, is a key aim of care, with 5‐aminosalicylate preparations being prescribed in most children with IBD. Whether early use of immunosuppressive drugs such as azathioprine is advantageous in the long‐term management of growth problems in IBD has not yet been established. Continuation of elemental or semi‐elemental nutrition as a nocturnal supplement, while eating normally during the day, has been found to be of some benefit in maintaining remission and improving linear growth, and can be considered for children suffering no or minimal symptoms but who have growth failure.43 Surgical intervention with resection of inflamed tissue may be necessary in patients with poorly controlled or severe disease. This is effective in inducing a remission, and can be associated with excellent post‐operative catch‐up growth.16 Severe growth failure, resistant to medical management, is an appropriate indication for surgery, particularly in the presence of localized disease.

Whatever strategy is used to induce disease remission, catch‐up growth will occur only when intervention is instituted before epiphyseal closure. It is essential that a diagnosis of growth failure is made promptly and adequate treatment is instituted to suppress disease activity and correct undernutrition.


Dr A. Ballinger is supported by the Wellcome Trust. The authors are grateful to the Crohn's in Childhood Research Association for financial support.


  • Address correspondence to Dr A.B. Ballinger, Digestive Diseases Research Centre, St Bartholomew's and The Royal London School of Medicine and Dentistry, Turner Street, London E1 2AT. e‐mail: A.B.Ballingermds.qmw.ac.uk