Ulcerative colitis (UC) and Crohn s disease (CD) are 2

Transcription

1 ORIGINAL ARTICLE Smoking and Inflammatory Bowel Disease: Trends in Familial and Sporadic Cohorts Jeffrey A. Tuvlin,MD,* Sania S. Raza, BA, Samuel Bracamonte, BA, Christina Julian, BA, Stephen B. Hanauer, MD, Dan L. Nicolae, PhD, Andrea C. King, PhD, and Judy H. Cho, MD Background: Ulcerative colitis (UC) and Crohn s disease (CD) result from genetic and environmental factors. Never smoking and formerly smoking increase the risk of UC, whereas smoking exacerbates the course of CD. We sought to define the age-dependent effects of smoking on the development of UC and CD in familial and sporadic cohorts. Methods: University of Chicago patients diagnosed with UC or CD between 1990 and 2002 were surveyed about their tobacco use relative to their diagnosis. Smoking trends were used to estimate age-dependent odds ratios and the attributable risks of smoking in the IBD cohort compared to in the general population. Results: One thousands and thirteen patients were included in the study: 245 with sporadic UC; 216 with sporadic CD; 249 with familial UC; and 303 with familial CD. Being an ex-smoker conferred an increased risk for UC in the age group in both the sporadic and familial cohorts, but not in the age group in the familial UC cohort. Furthermore, there was no difference in tobacco use between patients with sporadic CD and the general population, although there was a significant increase in smoking in younger patients with familial CD. Conclusions: Ex-smokers make up an increasing percentage of older patients diagnosed with UC, accounting for more than 35% of the attributable risk of late onset ( 45 years) UC and a large component of the second peak in diagnosis. Current smokers account for a large percentage of patients diagnosed at a younger age with familial CD but not with sporadic CD. Families with IBD should be counseled that early tobacco use significantly increases the risk of developing CD or, if an ex-smoker, UC at a young age. (Inflamm Bowel Dis 2007;13: ) Key Words: ulcerative colitis, Crohn s disease, smoking Received for publication July 27, 2006; accepted October 21, From *Midwest Gastroenterology Associates, Louisville, Kentucky; Southern Illinois University School of Medicine, Springfield, Illinois; University of Chicago, Chicago, Illinois; and West Virginia University, Morgantown, West Virginia. Reprints: Judy Cho, MD, Yale University, Section of Digestive Diseases, 300 Cedar St., S155A, New Haven, CT ( judy.cho@yale.edu). Copyright 2007 Crohn s & Colitis Foundation of America, Inc. DOI /ibd Published online 7 March 2007 in Wiley InterScience ( wiley.com). Ulcerative colitis (UC) and Crohn s disease (CD) are 2 diseases along a clinical spectrum of inflammatory bowel diseases (IBD). Although clinically and pathologically different entities, these 2 conditions have many similarities in epidemiology, clinical manifestations, and therapy. However, an intriguing difference is that tobacco seems to have opposite effects on UC and CD. There is a general consensus in the literature that the prevalence of UC has increased among nonsmokers and ex-smokers, whereas the results on the relation of CD to smoking status are less clear-cut. Studies have shown that the relative risk of developing UC is highest among ex-smokers, followed by lifelong nonsmokers, with a decreased risk in current smokers The onset of colitis after smoking cessation tends to be within the first 3 to 7 years after cessation. 7,11 Active smokers with UC also show a tendency toward decreased use of corticosteroids, a decreased colectomy rate, 15,17 and a decrease incidence of pouchitis after restorative proctocolectomy. 16 The data for CD are less clear. Most studies have shown that smoking has deleterious effects on the course of CD including greater recurrence after surgical and endoscopic procedures, more operations and complications, and reduced quality of life A clear association has also been found between tobacco use and CD location, most notably ileal disease. 9,19,20 There is a complex interplay between genetic and environmental factors in the development of IBD. Seventy-five percent of families in which more than 1 relative has IBD are concordant for disease type (all affected members have CD or all have UC). 21 It is likely that there are many susceptibility genes, some common to both CD and UC, but others specific to only one An example of a gene linked specifically to CD is Nod2/CARD15, in the pericentromeric region of chromosome ,29 Researchers also found that in multiply affected families (families in which more than 1 individual has UC or CD), smokers tend to develop CD and nonsmokers and ex-smokers tend to develop UC. 30 One study found that in 21 of 23 sibling pairs discordant for smoking status (a smoker and a nonsmoker), the smoker developed CD and the nonsmoker developed UC. 31 Several studies have shown sporadic and familial IBD to be distinct in their relative genetic and environmental contributions to clinical phenotype and therapeutic re- Inflamm Bowel Dis Volume 13, Number 5, May

2 Tuvlin et al Inflamm Bowel Dis Volume 13, Number 5, May 2007 sponse. 32 To date, the differences between the sporadic and familial subtypes of IBD in the effects of smoking tobacco on the development of UC and CD has not been studied. The present study compared the age-dependent effects of tobacco use in patients with sporadic IBD with those in patients with familial IBD and compared these effects with those in the general population. Age at onset of smoking, duration of tobacco use, and pack-years were used as interacting variables. We hypothesized that as a strongly associated environmental factor of IBD, smoking would have significantly different effects in the development of sporadic and familial IBD. Specifically, we expected tobacco to have a greater impact on sporadic IBD (especially late-onset UC) than on familial IBD, which is likely to be more affected by genetic influences. Age, duration of smoking, and pack-years were used to assess a threshold of tobacco use, if present. We hypothesized that increased pack-years and increased duration of smoking would contribute to an increased risk of IBD. Our results suggest that cessation of cigarette smoking is likely to account for a large percentage of the second age peak for UC, which is in the fifth and sixth decades of life, especially for sporadic cases. In sporadic CD, however, smoking contributes only minimally, whereas it doubles the risk of developing familial CD. MATERIAL AND METHODS Patient Population A detailed history of tobacco use was obtained from patients in 4 groups (defined below): a random cohort of patients with UC (sporadic UC), a random cohort of patients with CD (sporadic CD), a cohort of UC patients who had at least 1 family member who also had UC (familial UC), and a cohort of CD patients who had at least 1 family member who also had CD (familial CD). The sporadic IBD cohorts comprised patients without a family history of IBD ascertained from the University of Chicago Hospitals IBD Registry, a record of all IBD patients seen at these hospitals. The registry listed 1628 individuals who had been diagnosed with IBD between 1990 and This range of years was chosen to allow an adequate sample size while reducing the chances for recall bias and inaccuracy of subject reports. Of these 1628 patients we were able to contact 1371, 600 of whom agreed to participate. One hundred and thirty-nine registry patients with a family history of IBD, defined as having a first- or second-degree biological relative with IBD (61 of 306 UC patients [19.9%] and 78 of 294 CD patients [26.5%]), were excluded from the sporadic cohorts, leaving 461 patients with sporadic IBD (245 patients with sporadic UC and 216 patients with sporadic CD). The familial IBD cohorts were composed of patients in the University of Chicago Hospitals IBD Genetics Study Database. Patients in these cohorts, by definition, had a family history of IBD. Although we attempted to ascertain all affected members in this study, we only included 1 randomly selected member from each family in the database. To be part of the UC familial cohort, patients had to have at least 1 biological relative with UC, and to be part of the CD familial cohort, patients had to have at least 1 biological relative with CD. Of the 639 families that met the inclusion criteria, information was obtained for 552. There were 249 familial UC patients and 303 familial CD patients included in the study. No patient was included in both the sporadic and familial cohorts. National self-reported tobacco data of age-matched individuals from the 1999 National Health Interview Survey made up the fifth group, the age-matched control group. Estimates from this cohort were used to estimate the agedependent odds ratios and population-attributable risks of smoking status of the studied cohorts relative to the general population in order to best separate population smoking trends from studied variables. Smoking Definitions Informed consent was obtained according to the guidelines of the University of Chicago institutional review board. Patients were classified as nonsmokers, ex-smokers, or current smokers using date of diagnosis as the reference point. Consistent with Centers for Disease Control criteria, 33 nonsmoker was defined as a patient who reported having smoked less than 100 lifetime cigarettes. Also classified as a nonsmoker at the time of diagnosis was any patient who might have initiated cigarette smoking after being diagnosed with IBD. Current smoker was defined as a patient who reported having smoked 100 or more lifetime cigarettes and having begun to smoke before the year of diagnosis. These patients reported no quit date or a quit date after the date diagnosed with IBD. Ex-smoker was defined as a patient who reported having smoked 100 or more lifetime cigarettes and whose year started smoking and year quit smoking were both before the year diagnosed with IBD. Patient who reported that the year begun smoking was the same as the year diagnosed with IBD (n 4) had their smoking trends in the 30 days before diagnosis evaluated. A patient who smoked 3 or more days a week during that time was considered a current smoker at the time of diagnosis. A patient who smoked less than 3 days per week during that time was considered a nonsmoker. Patients who reported quit and diagnosis dates in the same year (n 27), reported smoking trends in the 30 days before diagnosis were evaluated. A patient who had smoked 3 or more days a week during this period was considered a current smoker. A patient who smoked less than 3 days per week during this period was classified as an ex-smoker. Averages were calculated for patients who reported considerable variability in smoking level. Also calculated were average pack-year use and, if an 574

3 Inflamm Bowel Dis Volume 13, Number 5, May 2007 Smoking and Inflammatory Bowel Disease TABLE 1. Smoking Status by Age Group of Persons 18 Years and Older in Ulcerative Colitis Cohorts Nonsmoker Ex-Smoker Current Smoker Odds Ratio* Population- Attributable Risk* UC sporadic (n 34) ( ) (n 130) ( ) 0.19 ( ) (n 54) ( ) 0.36 ( ) 65 (n 14) ( ) Total 232 UC familial (n 62) ( ) (n 108) ( ) 0.2 ( ) (n 23) ( ) 65 (n 9) ( ) Total 202 GP *OR and PAR are for ex-smoking as the risk factor in CD cohorts versus in the general population (GP), with 95% confidence intervals calculated conservatively assuming population data obtained from 1000 individuals per age group. PAR calculated only if OR was statistically significant. ex-smoker, number of years since smoking cessation. Packyears was calculated as the number of packs of cigarettes smoked per day the number years of smoking. The number of years since smoking cessation was calculated as year of diagnosis year quit smoking. Data and Statistical Analyses Consistent with the National Health Interview Survey age groups, the 4 cohorts were divided into age-based categories according to reported age at time of diagnosis. The 5 age subgroups in each cohort were: ( 1 ) 18 years, ( 2 ) years, ( 3 ) years, ( 4 ) years, and ( 5 ) 65 years. The percentages of nonsmokers, ex-smokers, and current smokers in each age group were calculated. Pack-years of tobacco use for current and ex-smokers were also calculated. In each cohort, each age group was compared to the general population using measurements of odds ratio and attributable risk. Population-attributable risk measures the excess risk associated with a risk factor. It is defined as the proportion of cases eliminated if the risk factor were nonexistent in the population and is calculated as (P Q)/(1 Q), where P is the frequency of the risk factor in the cases, and Q is the frequency of the risk factor in the population. 34 For UC, the risk is ex-smoking, and for CD it is current smoking. The 95% confidence intervals (CIs) were conservatively calculated assuming the population data were obtained with 1000 individuals per group. RESULTS A total of 1013 patients were included in the study (UC sporadic, n 245; CD sporadic, n 216; UC familial, n 249; CD familial, n 303), of which 494 were UC patients and 519 were CD patients. There were 461 sporadic patients and 552 familial patients. Tables 1 and 2 show the smoking habits of the UC and CD cohorts by age at diagnosis (18 years and older). These data were compared to smoking trends in the general population. The calculated odds ratios (ORs) and population-attributable risks (PARs) are shown using ex-smoking as the risk factor for the UC cohorts and current smoking as the risk factor for the CD cohorts. Ulcerative Colitis Cohorts Table 1 lists the smoking trends for the sporadic and familial UC cohorts by age at diagnosis as compared to the general population. Two hundred and forty-five patients were in the sporadic UC cohort, 95% of whom (232 patients) were older than age 18. The calculated odds ratios and populationattributable risks shown in Table 1 compare the ex-smoker subgroup to the general population, with ex-smoking as the risk factor. The percentages of ex-smokers in the 18 24, 25 44, and age groups in the sporadic UC cohort were 11.8%, 31.5% and 55.6%, respectively. Although the percentage of ex-smokers increased with increasing age in the general population, the percentage of ex-smokers in the sporadic UC 575

4 Tuvlin et al Inflamm Bowel Dis Volume 13, Number 5, May 2007 TABLE 2. Smoking Status by Age Group of Persons 18 Years and Older in Crohn s Disease Cohorts Nonsmoker Ex-Smoker Current Smoker Odds Ratio* Population- Attributable Risk* CD sporadic (n 42) ( ) (n 109) ( ) (n 38) ( ) 65 (n 4) Total 193 CD familial (n 100) ( ) (n 97) ( ) 0.20 ( ) (n 21) ( ) 65 (n 6) ( ) Total 224 GP *OR and PAR are for current smoking as the risk factor in CD cohorts versus in the general population (GP), with 95% confidence intervals calculated conservatively assuming population data obtained from 1000 individuals per age group. PAR calculated only if OR was statistically significant. cohort increased disproportionately, with a trend toward increased risk in the age group and statistically significant odds ratios of 2.5 and 2.9 for the and age groups, respectively. The contribution of ex-smoking to the development of late-onset sporadic UC (45 64 age group), as measured by the population-attributable risk, was 36%. The familial UC cohort comprised 249 patients, 81% of whom (202 of 249) were older than 18 years of age. In this cohort, as in sporadic UC, the percentage of ex-smokers increased from 4.8% to 28.7% and 30.4% in the 18 24, 25 44, and age groups, respectively. Although the same trend was observed in the general population, the percentage of ex-smokers in the age group was significantly higher in familial UC (28.7%) than in the general population (15.7%), with an odds ratio of 2.2 and a PAR of 20%. Because of the small number of patients diagnosed after 45 years of age, similar significance was not observed in the familial UC cohort. Crohn s Disease Cohorts Table 2 lists the smoking trends for the sporadic and familial CD patients by age at diagnosis as compared to the general population. The sporadic cohort included 216 patients, of whom 193 (89.4%) were older than 18 years. There was a peak in incidence of CD in the age group (109 of 216, 51%). The data on smoking status from the sporadic CD cohort and from the general population were strikingly similar, with no significant differences in smoking status in any age group. In the familial CD cohort, 224 of 303 patients were diagnosed at age 18 or older. Only a small percentage of patients were diagnosed after age 45 (27 of 303, 8.9%), unlike in the sporadic CD cohort, in which 42 of 216 patients (19.4%) were diagnosed after age 45. Of significance in this cohort was the percentage of current smokers between the ages of 25 and 44 (42.3%), which was higher than that of comparably aged people in the general population (27.5%), corresponding to an OR approaching 2.0 and a PAR of 20%. Pack-Year Effect on UC Cohort Data on pack-years of smoking of ex-smokers with sporadic UC are shown in Table 3. The sporadic cohort was examined because of the greater overall effect of ex-smoking in this cohort. Average pack-years of smoking increased with age at diagnosis, from an average of 0.3 pack-years in the 18 age group to 62.3 pack-years in the 65 age group. The possible contribution of pack-years to the increased risk of UC in older ex-smokers was evaluated by the percentages of ex-smokers diagnosed with UC at various pack-year amounts (Fig. 1). In this cohort, 33% of patients had less than 10 pack-years of tobacco use, and 59% of patients had less than 20 pack-years of tobacco use. The number of years since 576

5 Inflamm Bowel Dis Volume 13, Number 5, May 2007 Smoking and Inflammatory Bowel Disease TABLE 3. Average Pack-Years of Tobacco Use of Ex-Smokers with Ulcerative Colitis Age Ex-Smokers in Age Group Pack-Years cessation of smoking also increased by age at diagnosis: 0.2 years for those 18 to 24 years old, 4.4 years for those 25 to 44 years old, 8.5 years for those 45 to 64 years old, and 24 years for those more than 65 years old. Pack-Year Effect on CD Cohort Average pack-years of tobacco use was compared in the familial and sporadic CD cohorts by age at diagnosis, and no significant differences were noted. Specifically, in the age group, average pack-years of smoking was 2.1 pack-years for the sporadic cohort and 3.3 pack-years for the familial cohort. In the age group, pack-year use was 36.2 and 37.9 pack-years for the sporadic and familial cohorts, respectively. Because tobacco use increased CD risk in those years old in the familial cohort but not in the sporadic cohort, we sought to determine if the pack-years of tobacco use was higher in the familial cohort. However, no significant differences were found in average pack-years of smoking; in fact, the trend was in the opposite direction (18.1 pack-years in the sporadic cohort versus 16.2 pack-years in the familial cohort). DISCUSSION Environmental factors moderate phenotypic expression of complex genetic diseases such as IBD. As understanding of various contributions to the ultimate clinical expression of a disease process broadens, both the classification and the treatment of many diseases may be altered. First-degree relatives of a patient with CD have up to times the risk of developing CD than does the general population, 33 suggesting that the relative genetic and environmental contributions of sporadic and familial IBD to producing a clinical phenotype and response to therapy may be distinct. The average age at onset of CD has been shown to be earlier in familial cohorts (22 years) than in sporadic cohorts (26.5 years). 35,36 The converse has also been found, that is, the earlier the age at onset, the more likely it is that there is a positive family history. 32,37 To date, the differential effect of smoking tobacco on the development of UC and CD according to whether it is the sporadic or the familial form of the disease has not been studied. In the present study, the age at diagnosis and smoking habits of large cohorts of CD and UC patients divided into sporadic and familial cohorts were assessed. Our results indicated that late-onset sporadic UC is very significantly a disease of ex-smokers, with higher percentages than those observed in the general population, and comprises a large FIGURE 1. Pack-years of tobacco use of ex-smokers who have sporadic ulcerative colitis. [Color figure can be viewed in the online issue, which is available at 577

6 Tuvlin et al Inflamm Bowel Dis Volume 13, Number 5, May 2007 component of the well-described second peak in incidence of UC. To further assess the effect of smoking on the development of late-onset UC, we examined the effects of interacting variables, such as age, pack-years of tobacco use, and duration of ex-smoking. We hypothesized that increased diagnosis of late-onset UC in ex-smokers is a pack-year effect. Table 3 shows, as expected, that the number of pack-years increases with age at diagnosis. However, as seen in Figure 1, 33% and 59% of patients had less than 10 and 20 pack-years of tobacco use, respectively. Although in part this reflects the age at risk of UC, these data suggest that there is not a safe level of tobacco use or a threshold that must be crossed before risk develops. The number of years between quitting and developing UC also increases with age at diagnosis. Although this could explain the late-onset UC peak, it is unlikely that the longer a patient is an ex-smoker, the more likely she or he will develop UC. Previous reports suggested that the first 2 to 7 years after quitting is the most likely time to develop UC. 37,38 Furthermore, in our population, 68% of the patients were diagnosed with UC within the first 7 years after quitting tobacco use. Of interest was the finding that in the group of those years of age, ex-smoking conferred increased risk for UC only in the sporadic cohort, not in the familial cohort (Table 1). This age group accounted for 23% and 11% of the total number of cases in the sporadic and familial UC cohorts. Taken together, these findings suggest there is a range of pathogenic contributions, from early-onset familial UC (greater genetic contribution) to late-onset sporadic UC resulting from ex-smoking (greater environmental contribution with lesser contribution from genetic factors). There was a significant difference in smoking status between patients with familial CD and the general population between 25 and 44 years old. The odds ratio for current smoking in familial CD was almost 2.0. There was not, however, a significant difference in pack-years between smokers with familial CD and those with sporadic CD, suggesting no heavier use of tobacco in the at-risk group of familial CD. This may partly explain the conflicting reports in the literature on tobacco s contribution to the development of CD that is, the effect is modest and most significant in families whose genetic predisposition is higher, and thus modest increases in disease liability could result in an observable effect. Taking all the results of the present study together, we have concluded that the relative contribution of smoking to the development of IBD differs between patients with familial IBD and those with sporadic IBD. Among the latter, being an ex-smoker (above that expected in the general population) accounts for a large percentage of late-onset UC, and being a smoker contributes little to the development of CD. Among patients with familial IBD, being a smoker doubles the risk of developing CD, and being an ex-smoker doubles the risk of developing UC in younger ( ) age groups. These findings have important implications for both understanding the epidemiology of IBD and for estimating environmental covariates in genetic linkage and association studies. Families with IBD should be counseled that early tobacco use significantly increases the risk of developing CD or, if an ex-smoker, of developing UC. REFERENCES 1. Lindberg E, Tysk C, Andersson K, et al. Smoking and inflammatory bowel disease. A case control study. Gut. 1988;29: Harries AD, Baird A, Rhodes J. Non-smoking: a feature of ulcerative colitis. Br Med J (Clin Res Ed). 1982;284: Benoni C, Nilsson A. Smoking habits in patients with inflammatory bowel disease. Scand J Gastroenterol. 1984;19: Benoni C, Nilsson A. Smoking habits in patients with inflammatory bowel disease. A case-control study. Scand J Gastroenterol. 1987;22: Boyko EJ, Koepsell TD, Perera DR, et al. Risk of ulcerative colitis among former and current cigarette smokers. N Engl J Med. 1987;316: Franceschi S, Panza E, La Vecchia C, et al. Nonspecific inflammatory bowel disease and smoking. Am J Epidemiol. 1987;125: Motley RJ, Rhodes J, Ford GA, et al. Time relationships between cessation of smoking and onset of ulcerative colitis. Digestion. 1987; 37: Smith MB, Lashner BA, Hanauer SB. Smoking and inflammatory bowel disease in families. Am J Gastroenterol. 1988;83: Calkins BM. A meta-analysis of the role of smoking in inflammatory bowel disease. Dig Dis Sci. 1989;34: Persson PG, Ahlbom A, Hellers G. Inflammatory bowel disease and tobacco smoke a case-control study. Gut. 1990;31: Motley RJ, Rhodes J, Kay S, Morris TJ. Late presentation of ulcerative colitis in ex-smokers. Int J Colorectal Dis. 1988;3: Jick H, Walker AM. Cigarette smoking and ulcerative colitis. N Engl J Med. 1983;308: Tobin MV, Logan RF, Langman MJ, et al. Cigarette smoking and inflammatory bowel disease. Gastroenterology. 1987;93: Lorusso D, Leo S, Misciagna G, et al. Cigarette smoking and ulcerative colitis. A case control study. Hepatogastroenterology. 1989;36: Mokbel M, Carbonnel F, Beaugerie L, et al. Effect of smoking on the long-term course of ulcerative colitis. Gastroenterol Clin Biol. 1998;22: Merrett MN, Mortensen N, Kettlewell M, et al. Smoking may prevent pouchitis in patients with restorative proctocolectomy for ulcerative colitis. Gut. 1996;38: Fraga XF, Vergara M, Medina C, et al. Effects of smoking on the presentation and clinical course of inflammatory bowel disease. Eur J Gastroenterol Hepatol. 1997;9: Cottone M, Rosselli M, Orlando A, et al. Smoking habits and recurrence in Crohn s disease. Gastroenterology. 1994;106: Lindberg E, Jarnerot G, Huitfeldt B. Smoking in Crohn s disease: effect on localisation and clinical course. Gut. 1992;33: Russel M, Volovics A, Schoon E, et al. Inflammatory bowel disease: is there any relationship between smoking status and disease presentation? European Collaborative IBD Study Group. Inflamm Bowel Dis. 1988;4: Binder V. Genetic epidemiology in inflammatory bowel disease. Dig Dis. 1998;16: Armuzzi A, Ahmad T, Ling KL, et al. Genotype-phenotype analysis of the Crohn s disease susceptibility haplotype on chromosome 5q31. Gut. 2003;52: Forabosco P, Collins A, Latiano A, et al. Combined segregation and linkage analysis of inflammatory bowel disease in the IBD1 region using severity to characterise Crohn s disease and ulcerative colitis. On behalf of the GISC. Eur J Hum Genet. 2000;8:

7 Inflamm Bowel Dis Volume 13, Number 5, May 2007 Smoking and Inflammatory Bowel Disease 24. Mirza MM, Fisher SA, King K, et al. Genetic evidence for interaction of the 5q31 cytokine locus and the CARD15 gene in Crohn disease. Am J Hum Genet. 2003;72: Giallourakis C, Stoll M, Miller K, et al. IBD5 is a general risk factor for inflammatory bowel disease: replication of association with Crohn disease and identification of a novel association with ulcerative colitis. Am J Hum Genet. 2003;73: Negoro K, McGovern DP, Kinouchi Y, et al. Analysis of the IBD5 locus and potential gene gene interactions in Crohn s disease. Gut. 2003;52: Rioux JD, Daly MJ, Silverberg MS, et al. Genetic variation in the 5q31 cytokine gene cluster confers susceptibility to Crohn disease. Nat Genet. 2001;29: Hugot JP, Chamaillard M, Zouali H, et al. Association of NOD2 leucinerich repeat variants with susceptibility to Crohn s disease. Nature. 2001; 411: Ogura Y, Bonen DK, Inohara N, et al. A frameshift mutation in NOD2 associated with susceptibility to Crohn s disease. Nature. 2001;411: Lee JC, Lennard-Jones JE. Inflammatory bowel disease in 67 families each with three or more affected first-degree relatives. Gastroenterology. 1996;111: Bridger S, Lee JC, Bjarnason I, et al. In siblings with similar genetic susceptibility for inflammatory bowel disease, smokers tend to develop Crohn s disease and non-smokers develop ulcerative colitis. Gut. 2002; 51: Colombel JF, Grandbastien B, Gower-Rousseau C, et al. Clinical characteristics of Crohn s disease in 72 families. Gastroenterology. 1996; 111: Tobacco Information and Prevention Source (TIPS). CDC s National Health Interview Surveys, selected years; United States: (Centers for Disease Control Office on Smoking and Health). Cigarette Smoking Among Adults United States, Lachin J. Biostatistical Methods The Assessment of Relative Risks. New York NY: Sons W; Gasche C, Scholmerich J, Brynskov J, et al. A simple classification of Crohn s disease: report of the Working Party for the World Congresses of Gastroenterology, Vienna Inflamm Bowel Dis. 2000;6: Polito JM, 2nd, Childs B, Mellits ED, et al. Crohn s disease: influence of age at diagnosis on site and clinical type of disease. Gastroenterology. 1996;111: Farmer RG, Michener WM, Mortimer EA. Studies of family history among patients with inflammatory bowel disease. Clin Gastroenterol. 1980;9: