点击显示 收起
1 From the Department of Pediatrics, Division of Rheumatology (JMB), Nephrology (BJF and MBL), Gastroenterology and Nutrition (ES, BSZ, and VAS), The Children's Hospital of Philadelphia, University of Pennsylvania School of Medicine, Philadelphia, PA, and the Department of Epidemiology and Biostatistics, University of Pennsylvania School of Medicine, Philadelphia, PA (JMB, JS, BJF, and MBL)
2 Supported in part by the American College of Rheumatology Research and Education Foundation (JMB), the General Clinical Research Center (M01RR00240), and the Nutrition and Growth Center at the Children's Hospital of Philadelphia. 3 Reprints not available. Address correspondence to JM Burnham, Room 1562 CHOP North, The Children's Hospital of Philadelphia, 3535 Market Street, Philadelphia, PA 19104. E-mail: burnhams{at}email.chop.edu.
ABSTRACT
Background: Crohn disease (CD) in children is associated with low body mass index (BMI), poor growth, and delayed maturation; alterations in lean and fat mass, however, are poorly characterized.
Objective: The objective was to quantify lean and fat mass in children and young adults with CD and in healthy control subjects, relative to height and pubertal maturation.
Design: This cross-sectional study assessed whole-body lean and fat mass by using dual-energy X-ray absorptiometry in 104 subjects with CD and in 233 healthy control subjects aged 4–25 y. Linear regression was used to determine the effect of CD on body composition and to generate sex-specific SD scores (z scores) for lean and fat mass relative to height.
Results: Subjects with CD had lower height-for-age and BMI-for-age z scores (P < 0.001 for both) than did control subjects. CD was associated with significant deficits in lean mass after adjustment for height, age, race, and Tanner stage (P = 0.003); deficits in fat mass were not observed. The mean (±SD) lean mass–for-height and fat mass–for-height z scores in the subjects with CD were –0.61 ± 0.92 and –0.04 ± 0.86, respectively. Within the control group, fat mass–for-height was positively correlated with lean mass–for height (r = 0.41, P < 0.0001); this association was absent in the subjects with CD.
Conclusions: Children and young adults with CD had significant deficits in lean mass but preserved fat mass, which is consistent with cachexia. Further research is needed to identify physical activity, nutritional, and antiinflammatory interventions to improve body composition in persons with CD.
Key Words: Crohn disease • body composition • cachexia • lean mass • fat mass
INTRODUCTION
Crohn disease (CD) in children and adolescents is characterized by gastrointestinal tract inflammation with malabsorption, nutritional deficiencies, anemia, pubertal delay, growth failure, and osteopenia. Low body mass index (BMI; in kg/m2) is a well-recognized complication of CD in children (1, 2). However, previous studies evaluating fat mass and lean mass in children and adults with CD were based on small numbers of subjects and yielded conflicting results. Some studies reported that the lower body weight in subjects with CD reflected significant fat mass deficits with sparing of lean mass (3–7), whereas others reported significant deficits in lean mass (2, 8–13). Furthermore, the relations between body composition and disease characteristics, such as CD activity and type of therapy, have not been examined in a large cohort of children and adolescents with CD.
In adults, lean mass deficits are associated with demonstrable morbidity, including loss of muscle strength, altered energy metabolism, and increased susceptibility to infections (14). Inflammatory cachexia, which is defined as loss of lean mass without loss of fat mass, has been described in rheumatoid arthritis (15) and has been attributed to muscle-active cytokines, increased resting energy expenditure (REE), and physical inactivity. Although malnutrition is a well-recognized complication of CD, the prevalence of cachexia has not been determined in children or adults.
In children and adolescents, muscle forces are critical determinants of bone mineral accrual (16, 17). We recently evaluated whole-body bone mineral content in children and young adults with CD and in healthy control subjects (18). In that study, CD was associated with significant deficits in whole-body bone mineral content relative to height and maturation. However, adjustment for lean mass eliminated the bone mineral content deficit in the CD group compared with the control group. Therefore, deficits in lean mass and fat mass may have distinct effects on bone accrual, physical function, susceptibility to infection, and quality of life in children with CD.
Analysis and interpretation of body-composition data in children with a chronic illness requires careful attention to sex-, maturation-, and race-related differences in lean and fat mass relative to body size. Dual-energy X-ray absorptiometry (DXA) provides precise and accurate measures of lean mass and fat mass in children and adults (19, 20). The objectives of the present study were to assess lean mass and fat mass in children and young adults with CD and in concurrent healthy control subjects and to identify risk factors for alterations in fat mass and lean mass in subjects with CD.
SUBJECTS AND METHODS
Study subjects
Persons aged 4–25 y with CD who were being treated at the Children's Hospital of Philadelphia or the Hospital of the University of Pennsylvania were eligible for the study. Diagnosis was confirmed by radiographic, histologic, and clinical information. Persons with other medical conditions unrelated to CD that potentially affect growth or body composition, such as underlying renal insufficiency, thyroid disease, or known growth hormone deficiency, were excluded. Lumbar spine and whole-body bone measures, vitamin D concentrations, growth, and maturation have been reported in these subjects (18, 21–24).
Healthy control subjects were recruited from general pediatric clinics in the surrounding community and through newspaper advertisements. Control subjects were excluded for any coexisting conditions known to affect growth, nutritional status, dietary intake, or development. The protocol was approved by the Institutional Review Board at the Children's Hospital of Philadelphia. Informed consent was obtained from the young adult participants and the parents or guardians of those participants aged <18 y. Assent was obtained from those younger than 18 y.
Crohn disease characteristics
Medical records were reviewed for age at disease onset and diagnosis; disease characteristics; medical, nutritional, and surgical interventions; and hematologic and biochemical values after diagnosis. Site of disease was classified as upper gastrointestinal tract only (ie, proximal to the colon), colon only, or both. History of laboratory abnormalities was defined as one or more values outside the normal age and sex reference range after the diagnosis of CD had been established. CD severity was assessed by using the Pediatric Crohn's Disease Activity Index (PCDAI), which is based on history (30%), physical examination (30%), laboratory data (20%), and height velocity (20%) (25, 26). PCDAI scores are categorized as follows: no disease activity (0–10), mild disease activity (11–30), and moderate to severe disease activity (>30). PCDAI scores were measured at the study visit. Additionally, a cumulative PCDAI score was assigned by averaging all prior PCDAI scores documented in the medical record.
Medications
Use of the following medications was documented: 6-mercaptopurine, sulfasalazine, mesalamine [as Pentasa (Shire US Inc, Newport, KY) or Asacol (Proctor & Gamble Pharmaceuticals, Cincinnati, OH)], metronidazole, calcium supplementation, and corticosteroid enemas. All doses of enteral and parenteral corticosteroids were noted and were converted to prednisone equivalents. Corticosteroid exposure was summarized as lifetime cumulative prednisone dose (g) and as cumulative mg/kg based on body weight at the time of each dose. Average doses during intervals of corticosteroid therapy were summarized as mg/d and mg · kg–1 · d–1. This study was conducted before the use of tumor necrosis factor inhibitors in pediatric CD.
Anthropometry and pubertal development
Weight and height were measured with a digital scale to the nearest 0.1 kg (Scaltronix, White Plains, NY) and with a stadiometer to the nearest to 0.1 cm (Holtain Ltd, Croswell, Crymych, United Kingdom), respectively. Age- and sex-specific SD scores (z scores) for weight, height, and BMI were calculated by using the National Center for Health Statistics 2000 Centers for Disease Control and Prevention growth data and the LMS method (27). Pubertal stage was assessed according to the method of Tanner by a single investigator at the time of the study visit (28).
Dual-energy X-ray absorptiometry
Whole-body DXA scans were performed by using a Hologic QDR 2000 bone densitometer (Hologic Inc, Bedford, MA) with a fan beam in the array mode. The measurements were performed by using standard supine positioning techniques and were analyzed to generate estimates of lean mass (kg) excluding bone, and fat mass (kg). The skull was excluded from measures of lean and fat mass (postcranial). The instrument was calibrated daily with a hydroxyapatite phantom. The precision of lean and fat mass measures made with the use of DXA is reported as 2–3% and 3–4%, respectively (19).
Laboratory studies
Blood samples were obtained from subjects with CD at the time of the study visit. Serum albumin (g/dL), total protein (g/dL), hemoglobin (g/dL), and erythrocyte sedimentation rate (mm/h) were measured (Clinical Laboratory, Children's Hospital of Philadelphia, Philadelphia, PA) by using standard techniques.
Statistical analysis
Analyses were conducted by using STATA 8.2 (Stata Corporation, College Station, TX). Two-sided tests of hypotheses were used, and P values < 0.05 were considered significant. Differences in means were assessed by using Student's t test if the data were normally distributed and Wilcoxon's rank sum test if the data were not normally distributed. Group differences in categorical variables were assessed by using the chi-square or Fisher's exact test, where appropriate.
The primary outcomes were postcranial fat mass and lean mass in kg. CD is associated with poor linear growth; therefore, these measures were adjusted for stature. To test for body-composition differences between the CD and control groups, natural log–transformed multivariable linear regression models for fat mass and lean mass relative to natural log–transformed height were adjusted for the following covariates that may confound this comparison: age, race (African American versus all others), and Tanner stage of pubertal maturation (stage 1 as the referent group, with indicator variables for Tanner stages 2 through 5 given a value of zero or one if absent or present, respectively). Given the known sex differences in body composition during growth, all results are presented stratified by sex. In healthy children and adolescents, greater fat mass was associated with greater lean mass for height (29). Therefore, a multiplicative interaction term was used to determine whether the relation between fat mass and lean mass differed in the CD group compared with the control group. The assumptions of the regression models were assessed via graphical checks, the Shapiro-Wilk test of normality of residuals, the Ramsey omitted variable test, and the Cook-Weisburg test for heteroscedasticity.
Because the outcome variable was log-transformed, the independent effect of CD in each multivariate model was summarized as the adjusted ratio of the outcome measure in the subjects with CD divided by the outcome measure in the control subjects, along with 95% CIs; note that these ratios have no units. The adjusted ratios and 95% CIs were calculated as the exponentiated estimate of the regression parameters.
To assess the effect of CD-specific effects on body composition within the subjects with CD, fat mass and lean mass were converted to sex-specific z scores relative to height. For example, a lean mass–for-height z score of –1.0 indicates a whole-body lean mass that is 1 SD below that of control subjects of the same height and sex. Data from the control subjects were used to derive the predicted value of lean or fat mass relative to height by using linear regression. The model for lean mass relative to height did not exhibit heteroscedasticity; therefore, the root mean square error served as the SD. The model for fat mass relative to height did exhibit heteroscedasticity; therefore, height-specific SDs were estimated by regressing absolute residuals on height (30).
Factors associated with deficits in lean mass–for-height and fat mass–for-height within the group of subjects with CD were identified by using simple logistic regression; body-composition z scores were considered low if less than –1.0 (equivalent to the 16th percentile). Factors found to be significantly (P 0.05) associated with low body-composition z scores were entered into a multivariable logistic regression model. The correlation between body-composition z scores and continuous variables was assessed by using Pearson product-moment estimates.
RESULTS
Subject characteristics and body composition
A total of 104 subjects with CD and 233 healthy control subjects completed the study. Subject characteristics are summarized in Table 1. Control subjects were significantly younger than the subjects with CD. The subjects with CD were predominantly white, which is consistent with the demographics of the disease, and had delayed maturation: within Tanner stages 2 and 4, subjects with CD were an average of 1.4 and 1.5 y older than the control subjects (P < 0.05 and < 0.01, respectively), adjusted for sex and race.
View this table:
TABLE 1. Characteristics of subjects with Crohn disease and of healthy control subjects
The subjects with CD had significantly lower height-for-age, weight-for-age, and BMI-for-age z scores than did the healthy control subjects (all P < 0.0001). The BMI z score distribution within the control subjects was consistent with a recent report of the US population (31). There were no sex differences in age, Tanner stage, height-for-age z score, or BMI-for-age z score within the control or the CD group.
Disease characteristics
Disease characteristics, medications, and laboratory values were available in 56–88% of the subjects with CD and are detailed in Table 2. Among the 88% with complete medication data, 90% had a history of treatment with corticosteroids. Treatment and disease characteristics were compared between the male and female subjects with CD. Males were exposed to corticosteroids for a greater duration than were females [median duration: 15 mo (range: 0–128 mo) compared with 8 mo (range: 0–79 mo); P = 0.01] and thus received a greater total dose over the course of their disease [median dose: 7.9 g (range: 0–74.0 g) compared with 6.2 g (range: 0–41.7 g); P = 0.01]. During the exposure to corticosteroids, the doses for each sex (mg/d and mg · kg–1 · d–1) were similar. Corticosteroid dose (total mg/kg) was inversely correlated with height-for-age z score (r = –0.36, P = 0.0005). Males were more frequently treated with Pentasa than were females (33% compared with 16%; P = 0.01) and were more likely to have a history of hypoalbuminemia (27% compared with 8%; P = 0.02).
View this table:
TABLE 2. Disease characteristics, medication exposure, and laboratory assessments in subjects with Crohn disease
Lean mass
Lean mass (kg) and unadjusted lean mass–for-age and lean mass–for-height z scores for the subjects with CD and the healthy controls are shown in Table 3. The initial assessment of lean mass was adjusted for age and race. The ratio for lean mass in subjects with CD compared with that in the controls was 0.85 (95% CI: 0.80, 0.90; P < 0.001) in males and 0.88 (95% CI: 0.83, 0.94; P < 0.001) in females. Subsequently, potential confounders were added to the model; these included height and Tanner stage. The final model is summarized in Table 4, which shows that the ratio of lean mass in subjects with CD compared with that in controls was 0.94 (95% CI: 0.91, 0.98) for males and females, adjusted for height, age, Tanner stage, and race. Therefore, CD is associated with a 6% reduction in lean mass, independent of stature, age, maturation, and race. In a similar model that excluded black subjects in both the control and CD groups, the same 6% deficit in males and females was identified. Given the predominance of Tanner stage 1 control subjects, we performed a similar analysis in the Tanner stage 2–5 participants, stratified by sex, excluding black subjects. In this model, an 8% lean mass deficit in males (P = 0.003) and females (P = 0.001) was noted.
View this table:
TABLE 3. Unadjusted body-composition measures in subjects with Crohn disease and in healthy control subjects
View this table:
TABLE 4. Multiple linear regression analysis for log (lean mass)
Assessment of lean mass as a sex-specific z score relative to height also showed significant deficits in the CD group, as shown in Figure 1. The mean (±SD) lean mass–for-height z scores in the male and female control subjects were 0.00 ± 1.00, by definition. The lean mass–for height z-scores were significantly lower in the CD subjects: –0.58 ± 0.87 (P = 0.0002) in males and –0.65 ± 1.03 (P = 0.0004) in females. When the z scores were assessed relative to age, the deficits were greater: –1.16 ± 1.19 in males and –0.98 ± 0.97 in females (both P < 0.0001); lean mass–for age z-scores were significantly lower than lean mass–for height z-scores in the subjects with CD (P < 0.0001).
FIGURE 1.. Lean mass–for-height and fat mass–for-height z scores in subjects with Crohn disease (CD) and in healthy control subjects. There was a significant deficit in sex-specific lean mass–for-height z scores in both male and female subjects with CD (
Fat mass
Fat mass (kg) and unadjusted fat mass–for-age and fat mass–for-height z scores for the subjects with CD and the healthy controls are shown in Table 3. An initial model evaluated fat mass adjusted for age and race. The ratio for fat mass in subjects with CD compared with controls was 0.82 (95% CI: 0.66, 1.02; P = 0.08) in males and 1.01 (95% CI: 0.81, 1.25; P > 0.2) in females. Subsequently, potential confounders were added to the model, including height and Tanner stage. The final model, stratified by sex, which is summarized in Table 5, failed to detect significant deficits in fat mass in males or females with CD. The ratios of fat mass in subjects with CD compared with the controls, adjusted for height, age, Tanner stage, and race, were 0.93 (95% CI: 0.75, 1.16) in males and 1.12 (95% CI: 0.91, 1.38) in females. Similar results were found in a model that excluded black subjects from both the control and CD groups. No fat mass deficits were noted in a model including only Tanner stage 2–5 subjects and excluding black subjects.
View this table:
TABLE 5. Multiple linear regression analysis for log (fat mass)
Similarly, expression of fat mass as a sex-specific z score relative to height did not reveal significant deficits in CD. The fat mass–for-height z scores in males and females with CD were –0.10 ± 0.89 in males and 0.06 ± 0.79 in females (both P > 0.2 compared with controls), as shown in Figure 1. Fat mass–for-age z scores were –0.22 ± 0.79 (P = 0.11) in males and –0.18 ± 0.75 (P > 0.2) in females with CD and were significantly lower than the fat mass–for-height z scores (P < 0.0001).
Lean and fat mass interactions
To evaluate the relations between lean mass and fat mass in CD and in the controls, the model assessing lean mass (Table 4) was tested for a CD-by–fat mass interaction. The interaction term was significant in males (P = 0.004) and females (P = 0.007). When this relation was explored in separate models for the controls and the CD patients, lean mass was positively associated with fat mass in male and female healthy controls (both P < 0.001). However, males and females with CD did not have greater lean mass with greater fat mass. This relation is detailed in Figure 2 and demonstrates the positive slope of lean mass–for-height z score relative to fat mass–for-height z score in the control subjects only. No significant correlation was observed between lean mass–for-height z score and fat mass–for-height z-score in the CD subjects.
FIGURE 2.. Lean mass relative to fat mass in subjects with Crohn disease (CD) and in healthy control subjects. There was a positive correlation between sex-specific lean mass–for-height z score and fat mass–for-height z score in healthy control subjects (r = 0.41, P < 0.0001) by Pearson correlation. This relation was absent in children and young adults with CD (r = 0.05, P > 0.2). With the use of multivariable linear regression to examine lean mass, a CD-by–fat mass interaction term was significant in both males (P = 0.004) and females (P = 0.007).
CD-specific factors and body composition
Having established group differences in lean mass between subjects with CD and controls, we designed univariate models to examine the association of lean mass–for-height z scores with CD-specific variables summarized in Table 2. Overall, 27% of males and 29% of females had lean mass–for-height z scores <–1.0. The factors found to be significantly associated with a lean mass–for-height z score less than –1.0 are listed in Table 6. These factors are indicative of moderate to severe CD and were co-linear, ie, were highly correlated, because the prescribed drug therapies and nutritional interventions depend on the degree and location of inflammation and any concomitant weight loss or growth failure. For example, use of Pentasa was associated with hypoalbuminemia (P = 0.05) and anatomical site of disease (P = 0.03), because Pentasa is generally used to target inflammation in the upper gastrointestinal tract. Those treated with Pentasa also received a greater mean dose of corticosteroids when measured in mg/d (23.8 ± 12.2 compared with 17.3 ± 11.8, P = 0.02) or mg · kg–1 · d–1 (0.6 ± 0.3 versus 0.5 ± 0.3, P = 0.04). Subjects with a history of hypoalbuminemia were more likely to have received nasogastric feeding (P = 0.02). A multivariable model showed that among the covariates identified in univariate models, mesalamine (Pentasa) therapy was the only statistically significant factor associated with low lean mass-for-height z score. Addition of corticosteroid dose to the multivariable model did not influence the effect of Pentasa.
View this table:
TABLE 6. Univariate and multivariate logistic regression analysis of factors associated with low lean mass–for-height z score in subjects with Crohn disease1
Considering lean mass–for-height z score as a continuous variable, lean mass correlated negatively with PCDAI at the study visit (r = –0.26, P = 0.04) but not the cumulative PCDAI (r = 0.02, P > 0.2). None of the corticosteroid measures were significantly correlated with lean mass–for-height z score. A similar approach showed no CD-specific factors associated with low fat mass–for-height z scores or with fat mass–for-height as a continuous variable. Specifically, measures of corticosteroid exposure were not correlated with fat mass–for-height z score.
DISCUSSION
These data showed significant lean mass deficits and normal fat stores in a large, clinically heterogeneous sample of children and young adults with CD. Lean mass was inversely correlated with the PCDAI, the most widely used estimate of disease activity in pediatric CD. Furthermore, the expected positive association between fat and lean mass observed in healthy control subjects was absent in this sample of subjects with CD. During normal growth and development, lean mass increases steadily, particularly during the pubertal growth spurt (32). As adiposity and body weight increase, additional muscle is required to retain normal function (33, 34). The independent effects of pubertal maturation on lean mass in males and females are confirmed in Table 4.
This pattern of lean mass deficits without fat deficits is termed cachexia (15). In contrast, wasting represents deficits in fat and lean mass, which are typically due to inadequate dietary intake. Roubenoff et al (35) showed that inflammatory cachexia in rheumatoid arthritis was the result of altered energy and protein metabolism (greater REE and protein catabolism and reduced total energy expenditure), accompanied by inflammatory cytokine production. The muscle-active cytokines in rheumatoid cachexia, including tumor necrosis factor , interleukin 1ß, and interleukin 6 (15), are also significantly elevated in CD (36). These cytokines stimulate protein degradation, inhibit myogenic differentiation (37, 38), and induce myoblast apoptosis (39).
Corticosteroids, which were used in 90% of the CD subjects in our study, may directly affect muscle and fat mass. Adiposity is a well-recognized complication of corticosteroid use (40). Studies in animal models have shown that corticosteroids increased myostatin, which is a negative regulator of skeletal muscle mass (41). However, corticosteroid-induced reductions in muscle-active cytokines and disease activity may counteract direct effects on muscle. We recently reported that high-dose corticosteroids resulted in marked obesity without lean mass deficits in childhood nephrosis (42). Vaisman et al (43) documented similar findings in renal transplantation. In that study, recently transplanted subjects experienced initial increases in fat mass, which were likely secondary to corticosteroid use. Lean mass subsequently increased, which may be attributable to the decreasing corticosteroid dose, improved nutrition, and the normal association between fat and lean mass, as described in Figure 2. Our data in CD suggested a negative correlation between corticosteroid dose and lean mass, but this was not statistically significant. This may have been due to the limitations of retrospective assessment of corticosteroid exposure; the cumulative dose did not capture fully the intra- and intersubject variability in corticosteroid regimens. Also, CD severity confounds the potential association between corticosteroid dose and muscle deficits; PCDAI was associated with lean mass deficits, and corticosteroids were administered to patients with more severe disease.
Pentasa therapy was associated with both a higher cumulative corticosteroid dose and lower lean mass. Pentasa is a formulation of 5-aminosalicylic acid that is released throughout the small intestine and colon. In contrast, Asacol, another 5-aminosalicylic acid compound, is delivered primarily to the colon. Asacol is used to treat colitis and Pentasa is used primarily for small-bowel disease. Both drugs limit mucosal inflammation with minimal systemic absorption (44). Therefore, it is unlikely that Pentasa directly affected body composition. Rather, Pentasa therapy is likely a marker of small-intestine disease. Small-intestine disease is associated with micronutrient deficiencies, such as deficiencies of zinc and selenium, which may compromise growth and muscle mass (45, 46). These micronutrients were not measured in this study.
Whereas inflammation and drug therapy may compromise muscle accrual directly through sarcoactive cytokines, indirect effects on REE, nutrition, and physical activity may contribute to cachexia in CD. Energy balance is largely determined by the difference between dietary intake and total energy expenditure. In children, total energy expenditure comprises mainly physical activity energy expenditure and REE, with a minor contribution of diet-induced thermogenesis and growth. Azcue et al (10) showed that, compared with anorexia nervosa subjects, children with CD fail to down-regulate REE with malnutrition. However, greater REE has not been consistently shown in adults with CD (9, 10, 13, 47–50). Elevated REE has been shown in children with cystic fibrosis, who are susceptible to chronic inflammation, malabsorption, and malnutrition (51, 52).
Lower levels of physical activity may contribute to lean mass deficits in CD. Children with CD may have decreased physical activity as a result of fatigue, muscle weakness, and chronic abdominal or musculoskeletal pain. Unfortunately, physical activity was not assessed in the present study. However, physical activity interventions have been shown to improve muscle mass and function in other chronic inflammatory conditions, such as rheumatoid arthritis and chronic kidney disease (53, 54). Additionally, we did not assess dietary intake in our study population. Total energy intake, along with the carbohydrate, protein, and fat composition of the diet, may affect energy balance and body composition. For example, decreased dietary intake of protein or increased protein catabolism may adversely affect lean mass (55). Elevated diet-induced thermogenesis has been reported in adults with inactive CD involving the small intestine, which may further contribute to a negative energy balance (6).
An accurate determination of body composition in CD is of critical importance given the therapeutic goals: to induce disease remission and support normal growth and nutritional status. The analyses presented here illustrate the importance of a concurrent control group, with consideration of the confounding effects of short stature, delayed maturation, and race. Studies documenting altered body composition in children with inflammatory bowel disease are few. Boot et al (2) studied body composition in 55 children with inflammatory bowel disease (22 with CD, 33 with ulcerative colitis) by using DXA and documented low lean mass–for-age (–1.04 ± 1.41) and fat mass–for-age (–0.64 ± 1.02) z scores. Similar to our population, height-for-age z score deficits were present (–0.54 ± 1.25), and pubertal delay was noted. Azcue et al (10) studied anthropometric indexes, bioelectrical impedance measures, and total body potassium in 24 children with CD and 22 controls. Intracellular and extracellular body water was estimated by use of deuterium dilution and the corrected bromide space. Lean mass deficits were variable and depended on both the measurement method and the expression of lean mass as either an unadjusted variable or as a percentage of body weight. Fat mass deficits were present, expressed as a percentage of body weight. The authors compared 20 CD subjects treated with corticosteroids or enteral nutrition alone over a 3-mo interval. Both groups had significant increases in body weight and lean mass with treatment. The increases in lean mass and height were greater in the enteral nutrition group, and there was a trend toward increased percentage body fat in the corticosteroid group (P = 0.07). In both of these studies, failure to adjust for height-for-age may have resulted in an overestimation of the lean mass and fat mass deficits. Our analyses showed that lean mass deficits relative to age are significantly greater than lean mass deficits relative to height.
In summary, in children and young adults with CD, inflammatory cachexia may be due to muscle-active cytokines; corticosteroids; macro- and micronutrient deficits (56, 57); endocrinologic factors, such as insulin-like growth factor 1 (IGF-1) deficiency (58); increased REE; and decreased physical activity. Little is known about the long-term clinical significance of cachexia during childhood. However, poor prenatal and postnatal nutritional status have been associated with adverse consequences on body composition, muscle strength, and mortality during the adult years (59, 60). One likely consequence of decreased lean mass in children and adolescents with CD is impaired bone accrual (18). We recently reported that this same cohort of children and young adults with CD had significant deficits in whole-body bone mineral content, and bone deficits were strongly associated with lean mass deficits. Osteopenia in CD may result in an increased risk of osteoporosis-related morbidity and disability throughout adulthood (61, 62). Future studies are needed to evaluate the efficacy of nutrition and physical activity interventions to improve lean mass in children and young adults with CD. In addition, studies are required to explore the effects of newer biological therapies on growth and body composition in pediatric CD.
ACKNOWLEDGMENTS
JMB designed and performed the statistical analysis, did the background research, and wrote the manuscript. JS provided statistical expertise and assisted in manuscript preparation. ES participated in study design, patient recruitment, and data collection. BJF assisted in the statistical analysis and manuscript preparation. BSZ provided critical expertise in the design of the study and analysis of the body-composition data and assisted in manuscript preparation. VAS led in study design and data collection and aided in manuscript preparation. MBL was instrumental in the data analysis and manuscript preparation. The authors had no conflicts of interest to report.
REFERENCES