Inflammatory bowel disease (IBD) is the umbrella term used to describe chronic intestinal diseases including Crohn disease (CD), ulcerative colitis (UC), and indeterminate colitis (IC). The etiology of IBD is multifactorial, resulting from complex genetic, environmental, and immunological interactions (1). Epidemiological data from before 1990 reported the incidence of IBD to be significantly less in black subjects compared with white subjects (0.04–0.45 vs 1.35–3.5/100,000) (2,3). However, more recent studies suggest that the incidence and prevalence of IBD among black subjects is higher than initially reported. A survey study performed in a large health maintenance organization in California reported prevalence rates of 43.6/100,000 for white subjects and 29.8/100,000 for black subjects (4). One pediatric IBD study from Atlanta used their cohort to estimate an incidence of IBD in black children in their region of 7 to 12/100,000 (5).
Despite the large number of reports in the medical literature related to IBD, there is a scarcity of articles focusing on racial differences in disease presentation, clinical course, and response to medical and surgical management among adults and especially among children (5–9). It remains unclear whether the spectrum of IBD in ethnic groups in the United States, such as African Americans, is different or similar to the white population—if it is different, this could have implications for IBD diagnosis and management. Additionally, clinical IBD trials have suffered from low rates of minority subject enrollment, making it difficult to assess potential racial or ethnic variations in response to standard or new biological therapies (10,11). To determine whether racial variability can be identified, we decided to compare the clinical phenotype and disease course of black and white children with IBD treated at our institution during a 10-year period.
PATIENTS AND METHODS
Medical records of black and white children with IBD treated at the Johns Hopkins Children's Center from January 1991 through December 2000 were reviewed. Data collected included demographics, age at diagnosis, presenting symptoms, duration of symptoms before diagnosis, extraintestinal manifestations, and family history of IBD. Racial and ethnic status was self-reported. All black patients in this study were born in the United States and were of non-Hispanic origin.
The time of diagnosis was defined as the date of the first endoscopic procedure that provided histopathological confirmation of IBD. The diagnosis of CD or UC was based on endoscopic, histopathological, and radiological studies. Patients were classified as having IC if the findings of endoscopic, histopathological, and radiographic studies did not clearly meet the criteria for CD or UC. CD behavior was divided into inflammatory, stricturing, and penetrating disease based on the Vienna classification (12). Perianal disease was defined by the presence of perianal fistulas, abscesses, cavitating fissures, or swollen symptomatic skin tags. Colonic involvement for UC and IC was categorized as proctitis, left-sided colitis, or pancolitis.
Length of clinical follow-up was defined from the time of endoscopic diagnosis to the last inpatient or outpatient encounter. Height, weight, body mass index (BMI) z-scores, erythrocyte sedimentation rate (ESR), hemoglobin level, and albumin level were calculated or obtained at diagnosis and at 12 months after diagnosis. Exposure to corticosteroids, immunomodulator therapy (6-mercaptopurine [6-MP], azathioprine, cyclosporine, and methotrexate), and infliximab were recorded. Surgeries were defined as “major” if they included colectomy, intestinal resection, ileostomy, and fistulectomy. They were classified as “minor” if they included drainage of abscess, debridement, skin tag removal, lysis of adhesions, appendectomy and gastrostomy placement, or were a routine follow-up procedure to one previously performed.
All of the analyses were performed using STATA version 8.0 (Stata, College Station, TX). Two-sided t tests for independent samples were used to analyze differences in continuous clinical and demographic variables between black and white patients. Categoric data were displayed in contingency table format and statistical associations between race and outcome were assessed by χ2 goodness-of-fit tests. In addition, odds ratios and corresponding 95% CIs were calculated for categoric outcomes to show the strength of association between race and these outcomes.
The medical records of 245 children with IBD were reviewed. Of these 58 (24%) were black and 187 (76%) were white (Table 1). Mean age (±SD) for black patients was 12.6 ± 4.1 years, and that for white patients was 11.6 ± 4.5 years. Mean time of symptom duration before diagnosis and mean follow-up time were similar in both groups. A family history of IBD was noted to be present in 68 of 187 white children (36.4%) compared with 10 of 58 black children (17.5%; P = 0.006). White children were noted to have a 2-fold greater chance of being diagnosed before the age of 6 years than black children (OR [95% CI] = 2.0 [0.64–8.21]; P = 0.2).
At initial diagnosis and at follow-up, there was no significant difference between the 2 racial groups in the percentage of children within each IBD diagnosis category. A total of 26 patients had their original diagnosis changed during the study period (8.6% of black patients and 11.2% of white patients), resulting in the following disease distribution at the conclusion of the study period: among black children, 64% had the diagnosis of CD, 14% UC, and 22% IC; among white children, 67% had CD, 14% UC, and 19% IC.
There were no differences between the 2 racial groups with regards to symptom presentation. The most common presenting complaints were abdominal pain, rectal bleeding, diarrhea, vomiting, weight loss, and growth failure. Regardless of race, more patients with CD had abdominal pain as their presenting symptom compared with patients who had UC (43.7% vs 10%; P < 0.001). More patients with UC presented with rectal bleeding than compared with patients who had CD (75% vs 21.2%; P < 0.01).
Crohn Disease Location and Behavior
There were no differences in disease location in patients with CD, and in both racial groups, pancolitis was the most common phenotype for UC and IC (Table 2). Among patients with CD, 17.6% of black patients compared with 10% of white patients had perianal disease (OR [95% CI] = 1.8 [0.48–5.73].
During the study period, 29.2% of black children with CD had a change in their disease behavior from the less complicated inflammatory type to the more complicated stricturing or penetrating behaviors, compared with 11.1% of white children (P = 0.05). This resulted in 51.3% of black patients with a complicated disease behavior compared with 27.4% of white patients (P = 0.006; Fig. 1). The median time to change to stricturing or penetrating disease was 34.6 months in black children (range, 23.3–59.2 months), compared with 27.4 months in white children (range, 6.8–72.4 months).
The most common extraintestinal manifestation that developed in black and white children was arthralgia, which was found in approximately one third of the entire population. Skin manifestations, including pyoderma gangrenosum and erythema nodosum, tended to be more common in black children than in white children (10.3% vs 2.1%; P = NS). Sclerosing cholangitis, uveitis, and hepatitis occurred in a small number of patients, less than 3% among black and white patients.
The mean ESR at presentation in the black patients with CD was 55 ± 31 mm/hour compared with 29 ± 20 mm/hour in white patients (P > 0.001). Only 7.7% of black children with CD had a normal ESR at presentation compared with 41.4% of white children with CD (P = 0.02). No significant difference was noted in ESR between groups for UC or IC. The number of patients with hemoglobin levels lower than 10 g/dL at initial diagnosis was similar between groups, but at 12 months after diagnosis, 22.5% of black children had a hemoglobin level lower than 10 g/dL compared with 4.3% of white children (P = 0.001). There was no significant difference in albumin levels between groups noted at presentation or during the follow-up period.
There were no differences between racial groups in mean z-score for weight and height at disease presentation or at 12 months after diagnosis. However, at initial presentation 20% of the black children with CD had a BMI z-score less than −2, compared with only 2.7% of white children with CD (P = 0.009). All of these patients had their BMI z-scores corrected to greater than −2 at 12 months after diagnosis.
Medical and Surgical Therapies
A total of 89.7% of black children were prescribed corticosteroids, compared with 77% of white children (P = 0.035; Fig. 2). More black children (24.1%) were also administered infliximab compared with white children (12.8%; P = 0.037). There were no significant differences between groups in the number of patients exposed to immunomodulator therapy such as 6-MP, azathioprine, cyclosporine, or methotrexate. However, white children were twice as likely as black children to receive 6-MP or azathioprine within the first 3 months after diagnosis (OR [95% CI] = 2.2 [0.65–9.37]). No statistically significant differences were noted in the number of major or minor surgical procedures for CD or UC between racial groups, although black children with CD were 1.5 times more likely to undergo a major surgical procedure compared with whites: 10 of 34 (29.4%) vs 19 of 103 (18.4%; OR [95% CI] = 1.5 [0.53–3.71]).
We characterized the clinical phenotype and disease course of black and white children with IBD studied at our institution during a 10-year period. Our findings show that the distribution of CD and UC, anatomic distribution of the disease, and symptom presentation are similar in African Americans and whites (5,13,14). However, our study suggests that racial variations may exist with regards to family history and disease phenotype.
White children had a greater prevalence of IBD family history compared with black children, which is consistent with the findings of other investigators (5,9). Although ascertainment bias may be a contributing factor in the differences in family history between groups, the finding may also be due to genetic differences and biological variation among races (15). No specific UC susceptibility gene has been identified; however, recent studies have identified the first CD susceptibility gene, NOD2/CARD15(16). Associations have been noted between NOD2/CARD15 mutations and certain CD phenotypes such as earlier age of onset and small bowel involvement as well as racial and ethnic variations (17,18). A significantly lower frequency of NOD2/CARD15 mutations occurs in black subjects, indicating that the typical polymorphisms within or near the leucine-rich repeat domain of NOD2/CARD15 associated with disease pathogenesis in white and Jewish patients is either not present or altered in this population (9,17). Additionally, specific amino acid polymorphisms of NOD2/CARD15 unique to black subjects, such as the Arg790Gln variant, have been observed (19).
Differences in disease presentation and course between the 2 racial groups were most evident in individuals diagnosed with CD. At the time of diagnosis, significantly more black children had higher ESRs and BMI z-scores less than −2 than the white children. At 12 months a low hemoglobin level (<10 g/dL) was more commonly found in black children. Correlations have been shown between ESR increase and intestinal inflammation detected by leukocyte scintigraphy (20). ESR has been associated with increased morbidity in autoimmune disease and high ESRs, and low hemoglobin levels have been used as surrogate markers for increased disease severity in IBD (21,22). A low BMI is widely recognized in children with CD and has been associated with increased morbidity in children with chronic illness (23–25).
The findings of previous studies in adults have been mixed with regards to racial differences in CD behavior. In our study the black children were more likely to develop complicating fistulizing or penetrating disease during the follow-up period. Two adult studies have reported penetrating disease to be common in black subjects (7,13). Others have found no racial differences in disease behavior; however, these studies were cross-sectional in design and did not mention what definitions were used to classify disease behavior (9,14). Clinical CD behavior does not remain static but appears to evolve over time, going from uncomplicated inflammatory disease to a more aggressive stricturing or penetrating state. It has been reported that nearly half of adults experience a change in disease behavior over a period of 10 years and more than 80% experience a progression to stricturing or penetrating disease over a period of 20 years (26,27). Thus, evaluating a patient at one particular time point may underestimate the prevalence of complicated CD behavior. Interestingly, in a multicenter retrospective record analysis of adults with IBD, Nguyen et al noted that after 5 years from disease diagnosis, black patients were more likely than white subjects to have stricturing disease (28). The fact that we had a relatively longer mean duration of follow-up than other studies (approximately 5 years) may have enabled us to identify changes in disease phenotype over time that would have otherwise not been evident. Whether black children have more severe disease at presentation or are more at risk for developing aggressive disease behavior over time compared with white children requires additional investigation for validation. If these differences indeed exist, then it is intriguing to consider whether they could be caused by as yet unidentified genetic polymorphisms that may increase an aggressive disease phenotype or decrease a protective phenotype. The recent report of an uncommon variant of the IL23R gene conferring strong protection against CD is reflective of the fact that a continuum from susceptibility to protective effects can occur with functional genetic variation (29).
Few studies have examined patterns of pharmacotherapy among black subjects with IBD. Using corticosteroid use as an indicator of disease severity in data collected during a 3-year period (1989–1991) at University of North Carolina hospitals, black subjects were found to be 3 times more likely to use prednisone than white subjects. However, in the published study that evaluated self-reporting by patients, the investigators found no significant differences in usage of 5-aminosalicylates, corticosteroids, or immunosuppressive therapy (14). Our black patients had a higher exposure to corticosteroids and to infliximab than whites. It may have been that the aggressive disease behavior noted in the black children led to the increased use of these 2 medications. It is also possible that the increased use of corticosteroids and infliximab in our black population was the result of physician preferences, referral bias, or center-specific differences in health care utilization. Although we did not find any racial differences in frequency of use of 6-MP, azathioprine, cyclosporine, and methotrexate, white patients were more than 2 times more likely to be prescribed 6-MP/azathioprine within 3 months after the initial diagnosis. Studies that evaluate IBD management are needed to help determine whether increasing efforts should be made to adequately represent racial and ethnic minorities in clinical trials. Almost all clinical trials in IBD suffer from low rates of minority subject enrollment. For example, 2 key multicenter trials evaluating the efficacy of the tumor necrosis factor–α antibody infliximab had 0% and 4% minority enrollment, respectively (10,30). The pediatric IBD trials by Markowitz et al and Bousvaros et al included no greater than 8% minority enrollment (11,31). Thus, past clinical trials in IBD have been unable to determine the influence of race on clinical response to medical therapy.
A limitation of our study is that the patient population was studied at a tertiary referral center, so our patients may represent a group of children with greater IBD acuity and complications. This would not necessarily explain the racial differences noted between groups. If anything, we would have predicted that white children followed at our center would have been more likely to show evidence of greater disease severity because this group includes the majority of our patients with IBD who are referred by gastroenterologists because of intractable or unresponsive disease. In contrast, the black children followed at our institution come from the surrounding community and are typically referred by their primary care providers. It may be speculated that a decreased index of suspicion regarding the diagnosis of IBD in black children and inability to access health care in a timely fashion may have contributed to the differences noted in disease manifestations.
In summary, our study contributes to the sparse literature on the manifestations of IBD in racial minority groups and suggests that there are racial differences in IBD family history, phenotype, and therapeutic management between black and white children. Given the current deficiencies in understanding racial and ethnic variations in IBD, larger studies are needed to delineate whether the reported differences are valid or whether racial disparities exist in access to medical care and in disease management.
1. Oliva-Hemker M, Fiocchi C. Etiopathogenesis of inflammatory bowel disease
: the pediatric perspective. Inflamm Bowel Dis 2002; 8:112–128.
2. Loftus EF Jr. Clinical epidemiology of inflammatory bowel disease
: incidence, prevalence and environmental influences. Gastroenterology 2004; 126:1504–1517.
3. Reddy SI, Burakoff R. Inflammatory bowel disease
in African Americans. Inflamm Bowel Dis 2003; 9:380–385.
4. Kurata JH, Kantor-Fish S, Frankl H, et al
. Crohn's disease among ethnic groups in a large health maintenance organization. Gastroenterology 1992; 102:1940–1948.
5. Ogundi SO, Ranson JA, Sullivan K, et al
. Inflammatory bowel disease
in African-American children living in Georgia. J Pediatr 1998; 133:103–107.
6. Paul H Jr, Barnes RW, Reese VE, et al
. Crohn's disease in black patients. J Natl Med Assoc 1990; 82:709–712.
7. Goldman CD, Kodner IJ, Fry RD, et al
. Clinical and operative experience with non-Caucasian patients with Crohn's disease. Dis Colon Rectum 1986; 29:317–321.
8. Deveaux PG, Kimberling J, Galandiuk S. Crohn's disease: presentation and severity compared between black and white patients. Dis Colon Rectum 2005; 48:1404–1409.
9. Kugathasan S, Loizides A, Babusukumar U, et al
. Comparative phenotypic and CARD15 mutational analysis among African American, Hispanic, and white children with Crohn's disease. Inflamm Bowel Dis 2005; 11:631–638.
10. Hanauer SB, Feagan BG, Lichtenstein GR, et al
. Maintenance infliximab for Crohn's disease: the ACCENT I randomised trial. Lancet 2002; 359:1541–1549.
11. Markowitz J, Grancher K, Kohn N, et al
. A multicenter trial of 6-mercaptopurine and prednisone in children with newly diagnosed Crohn's disease. Gastroenterology 2000; 119:895–902.
12. Gasche CSJ, Brynskov J, D'Haens G, et al
. A simple classification of Crohn's disease: report of the Working Party for the World Congress of Gastroenterology, Vienna 1998. Inflamm Bowel Dis 2000; 6:8–15.
13. Simsek H, Schuman BM. Inflammatory bowel disease
in 64 black patients: analysis of course, complications and surgery. J Clin Gastroenterol 1989; 11:294–298.
14. Straus WL, Eisen GM, Sandler RS, et al
. Crohn's disease: does race
matter? Am J Gastroenterol 2000; 95:479–783.
15. Cummings JR, Jewell DP. Clinical implications of inflammatory bowel disease
genetics on phenotype. Inflamm Bowel Dis 2005; 11:56–61.
16. Hugot JP, Chamaillard M, Zouali H, et al
. Associate of NOD2 leucine-rich repeat variants with susceptibility to Crohn's disease. Nature 2001; 411:599–603.
17. Bonen DK, Cho JH. The genetics of inflammatory bowel disease
. Gastroenterology 2003; 124:521–536.
18. Economou M, Trikalinos TA, Loizou KT, et al
. Differential effects of NOD2 variants on Crohn's disease risk and phenotype in diverse populations: a metaanalysis. Am J Gastroenterol 2004; 99:2393–2404.
19. Bonen DK, Nicolae DL, Moral T, et al
. Racial differences in NOD2 variation: characterization of NOD2 in African-Americans with Crohn's disease [abstract]. Gastroenterology 2002; 122(Suppl):A-29.
20. Charron M. Temporal correlation of the erythrocyte sedimentation rate with Tc-99m WBC uptake on children with inflammatory bowel disease
. Clin Nucl Med 2006; 28:905–907.
21. Vila LM, Alarcon GS, McGwin JG, et al
. systemic lupus erythematosus in a multiethnic cohort (LUMINA): XXIX. Elevation of erythrocyte sedimentation rate is associated with disease activity and damage accrual. J Rheumatol 2005; 32:2150–2155.
22. Spivak W, Sockolow R, Rigas A. The relationship between insurance class and severity of presentation of inflammatory bowel disease
in children. Am J Gastroenterol 1995; 90:982–987.
23. Burnham JM, Shults J, Semeao E, et al
. Body-composition alterations consistent with cachexia in children and young adults with Crohn disease
. Am J Clin Nutr 2005; 82:413–420.
24. Weinstein TA, Levine M, Pettei MJ, et al
. Age and family history at presentation of pediatric inflammatory bowel disease
. J Pediatr Gastroenterol Nutr 2003; 37:609–613.
25. Zhang Z, Lai HJ. Comparison of the use of body mass index percentiles and percentage of ideal body weight to screen for malnutrition in children with cystic fibrosis. Am J Clin Nutr 2004; 80:982–991.
26. Louis E, Collard A, Oger AF, et al
. Behaviour of Crohn's disease according to the Vienna classification: changing pattern over the course of the disease. Gut 2001; 49:777–782.
27. Cosnes J, Cattan S, Blain A, et al
. Long-term evolution of disease behavior of Crohn's disease. Inflamm Bowel Dis 2002; 8:244–250.
28. Nguyen GC, Torres EA, Regueiro M, et al
. Inflammatory bowel disease
characteristics among African Americans, Hispanics, and non-Hispanic whites: characterization of a large North American cohort. Am J Gastroenterol 2006; 101:1012–1023.
29. Duerr RH, Taylor KD, Brant SR, et al
. A genome-wide association study identifies IL23R as an inflammatory bowel disease
gene. Science 2006; 314:1461–1463.
30. Targan SR, Hanauer SB, van Deventer SJH, et al
. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor α for Crohn's disease. N Engl J Med 1997; 337:1029–1035.
31. Bousvaros A, Guandalini S, Baldassano RN, et al
. A randomized, double-blind trial of Lactobacillus GG versus placebo in addition to standard maintenance therapy for children with Crohn's disease. Inflamm Bowel Dis 2005; 11:833–839.