Juvenile onset inflammatory bowel disease: height and body mass index in adult life

BMJ 1994; 308 doi: (Published 14 May 1994) Cite this as: BMJ 1994;308:1259
  1. A Ferguson,
  2. D M Sedgwick
  1. Gastrointestinal Unit, Department of Medicine, Western General Hospital, Edinburgh EH4 2XU
  1. Correspondence to: Professor Ferguson.


    Objective : To establish the frequency of permanent growth failure in juvenile onset inflammatory bowel disease.

    Design : Measurement of height and weight in a geographically based cohort at a mean of 14 (range 5.2-29.5) years after diagnosis. Comparison with data from surveys of British adults in 1980 and 1987.

    Setting : NHS hospitals throughout Scotland. Subjects - 105 Children admitted to hospital during 1968-83 who fulfilled diagnostic criteria for Crohn's disease or ulcerative colitis and lived in specified regions. 87 were aged over 18 and living in Britain at follow up.

    Main outcome measures : Height, weight, body mass index, and sexual maturity.

    Results : All patients were sexually mature. 67 of the 70 patients examined were of normal height, and three women with Crohn's disease were abnormally short. Weight and body mass index were normal in all patients with ulcerative colitis. Patients with Crohn's disease had significantly lower weight than those with ulcerative colitis (men 66.8 (9.5) kg v 78.4 (13.8) kg, P=0.04; women 51.5 (8.2) kg v 63.0 (12.1) kg, P<0.02) irrespective of disease activity. Body mass index was also significantly lower than the normal distribution (P<0.01). Growth retardation was not mentioned as a problem for any of the 17 patients interviewed only by telephone.

    Conclusions : Despite growth retardation in the teenage years most young people with inflammatory bowel disease will eventually achieve normal height. Reasons for lower weight in patients with Crohn's disease remain to be established.

    Clinical implications

    • Clinical implications

    • Growth retardation is common in children with inflammatory bowel disease, particularly Crohn's disease

    • In this study 67 of 70 children with Crohn's disease or ulcerative colitis were of normal height as adults, mean of 14 years after onset of disease

    • Patients with Crohn's disease had significantly lower weight than the normal population

    • All patients were sexually mature

    • Children and teenagers with inflammatory bowel disease should have height and weight measured regularly and sexual development recorded


    A study of a geographically based sample of 105 patients with juvenile onset inflammatory bowel disease identified through the Scottish hospitals inpatients statistics system confirmed previous reports of high early morbidity.*RF 1-5* Many children with Crohn's disease were short and underweight for their age, and no improvement was found between diagnosis and follow up a mean of 7.2 years later.2 Girls were more likely to be underweight than boys. In contrast, the height and weight centiles were normal in children with ulcerative colitis.2 A further review of this series of patients has been undertaken to establish their health in young adult life.6,7 We report here data on height and weight.

    Subjects and methods

    Our original sample of 105 patients comprised patients admitted to Scottish NHS hospitals between 1968 and 1983. Criteria for inclusion were a diagnosis of Crohn's disease (ICD 555.0-555.9) or ulcerative colitis (ICD 556.0); fulfilment of diagnostic criteria for Crohn's disease or ulcerative colitis; development of symptoms of inflammatory bowel disease before the age of 17; resident in Lothian (mainly urban) or Highland (rural) regions or, for Crohn's disease only, being in a 10% sample of cases from the rest of Scotland.2 Subgroup analyses showed no differences between the 10% sample and the Lothian and Highland subgroups, and all data were combined.

    We traced patients through hospital records, parents, general practitioners, and NHS regional records. With the agreement of the hospital consultant or general practitioner we wrote to patients and invited them to attend for clinical assessment. We evaluated the current health of those unwilling to attend by a telephone interview with either the patient or a parent. The protocol was approved by the Lothian medicine and clinical oncology ethics of research subcommittee.

    Full details of the clinical assessments have been reported.6 We calculated the Crohn's disease activity index8 for patients with Crohn's disease and ulcerative colitis using a symptom diary completed by the patients for the seven days before the interview. In patients who had a stoma (15 with Crohn's disease, seven with ulcerative colitis) the number of loose stools was taken as the number of times the patient emptied liquid faeces from the ileostomy bag.

    Clinical examination included assessment of pubertal status and measurements of height and weight. To measure height we asked patients to stand, bare feet together, with the back of their heels against a wall, head forward, and chin tilted so that their eyes were level with their ears. We then measured them with a wall mounted ruler.

    Values for heights and weights of British men and women aged 20-34 years and for percentage distribution of body mass index in Scottish men and women were obtained from an extensive survey carried out in 1980.9 More recent values were also taken from a smaller survey conducted in 1987.10 Body mass index was calculated as weight (kg)/(height (m)<sup>2</sup>).

    Data are presented throughout as mean plus 1 SD. We used the two tailed t test to compare patients with Crohn's disease and ulcerative colitis. Analysis of variance was used to compare distribution of body mass index in our patients with that in the general population.


    Of the original cohort of 105 patients with inflammatory bowel disease, six had died, seven had emigrated, four were still children, and one did not, in retrospect, fulfil age criteria for the study. The remaining 87 were all traced and were alive in January 1991. We examined 70 patients (50 with Crohn's disease, 20 with ulcerative colitis) and made telephone contact with the remaining 10 patients with Crohn's disease and seven with ulcerative colitis.

    Patients who had clinical assessment

    Among the patients we examined, general morbidity, as reflected by hospital admissions, was similar for those with Crohn's disease (mean 6.6 admissions, 102 inpatient days) and those with ulcerative colitis (mean 6.4 admissions, 93 inpatient days) (table I) and for men and women. For Crohn's disease, the patients examined and those contacted only by telephone were very similar. For ulcerative colitis, however, the patients who declined to attend for clinical examination were predominantly men with less extensive disease and fewer hospital admissions (table I).6


    Demographic and clinical details of patients with inflammatory bowel disease who were examined or interviewed by telephone

    View this table:

    Sixty eight of the 70 patients reported that their symptoms and general health at the time of the clinical interview reflected their clinical state for the previous six months or longer; one man with Crohn's disease had recently developed obstructive symptoms after a long remission, and one man had a recent relapse of ulcerative proctitis. One woman with Crohn's disease had intestinal failure and was receiving home total parenteral nutrition.

    Figure 1 shows the Crohn's disease activity index for the patients. Ten of 22 women with Crohn's disease had an activity index above 150 compared with only six of the 28 men (P<0.05). The Crohn's disease activity index was above 150 in only two of the 20 patients with ulcerative colitis.

    FIG 1
    FIG 1

    Crohn's disease activity index at mean of 14 years after diagnosis in 50 patients with juvenile onset Crohn's disease and 20 with juvenile onset ulcerative colitis.

    Table II and figures 2 and 3 show the heights and weights of patients with Crohn's disease and ulcerative colitis. The height of patients with Crohn's disease and ulcerative colitis was similar to that of normal adults, and we found no difference between patients with Crohn's disease and ulcerative colitis or men and women (P>0.1), although there were three abnormally short women with Crohn's disease. The weight of men (P=0.04) and women (P<0.02), with Crohn's disease was significantly lower than that of those with ulcerative colitis.


    Data on height and weight for Scottish men and women with juvenile onset inflammatory bowel disease and for normal British population in some age group9,10

    View this table:
    FIG 2
    FIG 2

    Height at clinical review in early adult life (mean 14 years after diagnosis) for 50 patient swith juvenile onset Crohn's disease and 20 with juvenile onset ulcerative colitis. Relative distributions of height for British men and women in same age group are also shown

    FIG 3
    FIG 3

    Weight at clinical review in, early adult life (mean 14 years after diagnosis) for 50 patient with juvenile onset Crohn's disease add 20 with juvenile onset ulcerative colitis. Relative distributions of weight for British men and women in same age group are also shown

    The three short women had no specific common features. One woman (height 1.41 m), aged 27 at follow up, had developed orofacial and anal Crohn's disease at age 10; she had had no surgery and her only recent symptoms were mouth lesions. At the time of her clinical assessment she weighed 63 kg. Another 27 year old woman (height 1.44m, weight 43.4 kg) had had Crohn's disease since the age of 12. It was complicated by duodenal ulceration with duodenal stenosis, and she had had four laparotomies with appendicectomy, three small bowel resections, vagotomy, and gastroenterostomy. She was the most seriously ill patient in the series and received home total parenteral nutrition. The third patient (height 1.45m, weight 33.6 kg) had developed small bowel and anal Crohn's disease at age 10. She was 19 years old at the time of the assessment and had recently had pan-proctocolectomy. Her height did not increase during a further two years of follow up.

    Table III shows the distribution of body mass index. Distribution of body mass index was similar to that of the normal population in men and women with ulcerative colitis but significantly skewed towards low values for body mass index in those with Crohn's disease (P<0.01).


    Distribution of body mass index in Scottish men and women and in patients with juvenile onset inflammatory bowel disease. Figures are numbers (percentages)

    View this table:

    Although relative weight for height is a factor used to calculate the Crohn's disease activity index,8 low body mass index was not related to disease activity. Among patients with Crohn's disease, body mass index ranged from 14.9 to 27.1 in men with active disease; from 16 to 25.6 in men with inactive disease; from 16 to 24.7 in women with active disease; and from 13.1 to 31.7 in women with inactive disease.

    Limited information on previous treatments had been collected, and we could not estimate lifetime doses of drugs. Systemic corticosteroids had been taken at some time by 19 of 28 men and 16 of 22 women with Crohn's disease (only one of the three short women) and by five of nine men and seven of 11 women with ulcerative colitis. At the time of assessment seven patients with Crohn's disease and one with ulcerative colitis were taking oral prednisolone 5-15 mg/day.

    Four patients with ulcerative colitis were obese. Corticosteroid treatment for chronically active disease may have contributed to this in a woman who weighed 85 kg, but the three men who weighed over 85 kg had never taken steroids.

    Information on patients interviewed by telephone

    Documentation of height and weight in the case records was poor.5 When we made telephone contact with the 17 patients who did not attend for interview we asked about general health and wellbeing but not about height. Two patients with Crohn's disease, a man and a woman, had chronic ill health. They had spent long periods in hospital and were being considered for home total parenteral nutrition. No patient or parent indicated short stature was a problem, and permanent growth retardation was not mentioned in the hospital records of any patient. Of the seven women with Crohn's disease, one was 1.54m at age 14 and two were normal height as adults (1.63 m and 1.65 m); height was recorded for one of the three men with Crohn's disease (1.72 m). No data on heights were available for the five men and two women with ulcerative colitis.

    Sexual development

    Records of growth and development were incomplete, and we did not attempt to assess the timings of growth spurt or the stages of puberty. However, when asked if they had developed at the same time as their peers, 11 of the 28 men with Crohn's disease and 13 of the 22 women reported delayed development; for ulcerative colitis the figures were two of nine men and three of 11 women.6 Differences between patients with Crohn's disease and ulcerative colitis were not significant. All patients had normal external genitalia, but vaginal examination was not done.

    The age at menarche was similar to that of the general population in women in the ulcerative colitis group, but several women with Crohn's disease who had not menstruated before the onset of symptoms reported considerably delayed menarche (fig 4).

    FIG 4
    FIG 4

    Age at menarche in 22 women with juvenile onset Crohn's disease and 11 with juvenile onset ulcerative colitis according to timing of onset of symptoms of inflammatory bowel disease


    Reports of growth and development in young people with inflammatory bowel disease consistently record retardation of sexual maturation and of growth.11 Reduced growth has been recognised even before the onset of symptoms,12 and growth spurts after surgery may be transient and unpredictable.*RF 13-16* Crohn's disease and ulcerative colitis cases are not differentiated in all reports, but generally it is accepted that growth delay is less of a problem in ulcerative colitis than Crohn's disease.

    Recent follow up data for several series of children with Crohn's disease have shown that although successful treatment may lead to resumption of linear growth rates, catching up in the teenage years is unusual.2,17,18 Many patients in these series, however, were still teenagers at the end of the study period, although it seems likely that some of them will be abnormally short in adult life.

    Sixty seven of the 70 patients we examined ultimately achieved a normal adult height. There were no common features in the three patients with permanent short stature, apart from being female and having Crohn's disease. Of the remaining 10 patients with Crohn's disease in the original cohort, four are known to be of normal height and for the others no comments on abnormal stature have been found in hospital records or in our contacts with them or their relatives.

    The patients with ulcerative colitis whom we examined had normal weight and body mass index, but weight and body mass index were lower than in the general population in men and women with Crohn's disease. Body mass index and Crohn's disease activity index were not significantly associated.

    American findings

    A study in American reported permanent impairment of growth in 14 of 38 (37%) patients with Crohn's disease and one of 10 with ulcerative colitis.19 The heights were judged in relation to American population norms (mean SD 1.725 (0.062)m for men; 1.639 (0.073) m for women), which are similar to published normal values for British adults and values in our patients. The heights of the New York patients with growth failure (men 1.649 (0.06)m; women 1.48 (0.096)m) were clearly shorter than those of our patients.

    Patients in the American and our series had similar clinical features such as age at onset and disease distribution. Our series was a geographically based sample, and patients were treated by many paediatricians, gastroenterologists, and surgeons whereas the American series was from a single specialist centre and may reflect a degree of referral bias. The most striking differences between the two series, however, was in the management strategies. The 15 American patients with impaired growth (14 with Crohn's disease) had a high rate of treatment with oral or intravenous nutritional supplementation (50%), oral corticosteroids (85%), and mercaptopurine or azathioprine (60%) and relatively low rates of surgery (38%). In our series more patients had had surgical treatment (57% at 7.2 year follow up, 78% at 14 years) and fewer had had second line medical treatment (72% oral corticosteroids, 18% azathioprine, 24% nutritional therapy).6

    Mechanisms of growth retardation

    The mechanisms which cause growth delay and pubertal arrest in teenagers may also cause low weight, irrespective of disease activity, in adults with Crohn's disease. Possible factors include growth hormone deficiency, impaired metabolic responses to growth hormone, low in take of energy and specific nutrients, loss of protein and blood through the gut, and increased metabolic rate.11 Resting metabolic rate and total daily energy expenditure were reported to be normal in patients with stable inflammatory bowel disease,20 and growth retarded children with inflammatory bowel disease generally have normal growth hormone responses.21 These patients, however, have low concentrations of insulin-like growth factor 1,22 and this has also been reported in malnourished adults with inflammatory bowel disease.23

    As intensive nutritional replacement can reverse growth failure, malnutrition is thought to be an important factor in growth retarded children with Crohn's disease.24 Investigation of malnutrition in adults will require careful recording not only of energy and nutrient intake and absorptive capacity but also of patterns of eating, appetite, and feeding behaviour. There are many potential approaches to the prevention and treatment of low weight in patients with inflammatory bowel disease, including nutritional therapy, immunosuppression, immunomodulation, exercise programmes, endocrine treatment, and intervention with appetite stimulants. Adults with ulcerative colitis who have normal weight and height might provide useful controls for comparisons with patients with low weight and low body mass index in future studies.

    When a chronic and essentially incurable disease is diagnosed in a child, many issues of prognosis and likely morbidity are raised. Parents of children with inflammatory bowel disease can be reassured that adult height and sexual development will usually be normal.

    We thank the clinicians who allowed us to examine their patients, and medical records officers in hospitals throughout Scotland. We acknowledge the help and advice of Dr J Clarke and his staff in the Common Services Agency and Mrs Jean Drummond in collating the data. This work is supported by the Crohn's in Childhood Research Association and the Edinburgh Intestinal Immunology Research Fund.


    View Abstract