Clinical predictors of inflammatory bowel disease in a genetically well-defined Caucasian population
© Kanaan et al; licensee BioMed Central Ltd. 2012
Received: 19 September 2011
Accepted: 23 January 2012
Published: 23 January 2012
Crohn's disease (CD) and ulcerative colitis (UC), the two main types of inflammatory bowel disease (IBD), are multifactorial conditions of unknown etiology. The objective of this study is to examine the combined gene-environment interactions influencing IBD susceptibility in a well-defined Caucasian cohort in rural mid-America.
Patients were diagnosed to have CD or UC using conventional radiologic, endoscopic, and/or histopathologic findings. Histological diagnosis was made by a single specialist gastrointestinal pathologist with a particular interest in IBD. Information regarding cigarette smoke exposure was obtained by administration of the Behavioral Risk Factor Surveillance System Survey (BRFSS) to all patients. Genomic DNA was extracted from peripheral blood leukocytes, and polymerase chain reaction (PCR) amplification and genotyping were performed for 11 Single Nucleotide Polymorphisms (SNP) in NOD2, IL23r, OCTN1 genes along with IGR.
Our cohort consists of 1196 patients: 435 controls, 485 CD patients, and 276 UC patients. Only patients with genotype data for at least 7 of 11 SNPs were included in our data analysis. The control groups for all 11 SNPs were in Hardy-Weinberg Equilibrium. In genotype-association SNP analysis, all NOD2 SNPs (rs5743293, rs2066844, rs2066845) and the IL23r SNP (rs11465804) showed a significant association to IBD (p < 0.03). A multiple gene-interaction analysis showed an association between NOD2 and IL23r with UC (p = 0.04). There were no associations between any OCTN1 and IGR SNPs and IBD in this cohort. A multivariable logistic regression analysis showed that female gender, "current" or "former" smoking status, family history of IBD, and NOD2 SNP minor alleles were associated with CD.
IBD remains to be challenging to properly diagnose, characterize, and treat. Our study proposes a combined genetic, phenotypic, and environmental approach in an attempt to better understand IBD. Previously demonstrated associations between OCTN1 and IGR and IBD were not confirmed.
KeywordsInflammatory bowel disease Crohn's disease Ulcerative Colitis SNP NOD2 IL23r OCTN1 IGR
Inflammatory bowel disease (IBD) is a chronic inflammatory disorder of the gastrointestinal (GI) tract, comprised of Crohn's disease (CD) and ulcerative colitis (UC). Over the past century, Northern Europe and North America have witnessed a significant rise in incidences of IBD . The inflammatory response in CD patients is described by transmural inflammation in any portion of the gastrointestinal tract while that of UC is usually limited to the mucosa and submucosa of the colon and rectum . Although the exact pathogenesis is not completely known in IBD, our current understanding suggests a disease etiology dependent upon a multifaceted interaction between genetic, environmental, and clinical factors.
Evidence of genetic factors implicated in this condition is supported by increased rates of IBD in monozygotic twins, and the ethnic differences in IBD frequency . In addition, different races and ethnic groups have different degrees of susceptibility to IBD . A number of studies have shown an association between IBD susceptibility and the nucleotide-binding oligomerization domain 2 gene (NOD2, also known as CARD15), interleukin-23 receptor gene (IL23r), organic cation transporter novel type 1 gene (OCTN1), and the intergenic region (IGR) variants [6–9]. NOD2 gene mutations have been extensively studied and described in the literature in terms of IBD susceptibility. Loss-of-function mutations in NOD2 gene appear to be the most significant for development of IBD, although the mechanism that increases disease susceptibly is poorly understood. One report suggests that NOD2 contributes a protective function in host defense that when compromised elicits a loss in immune bacterial recognition . The NOD2 gene on chromosome 16 has been specifically implicated in susceptibility to CD with 30-50% of Caucasian CD patients having variants of this gene . The three NOD2 variant single nucleotide polymorphisms (SNPs) (rs2066844, rs2066845, and rs5743293) have been strongly associated with clinical presentation of inflammatory bowel disease [1, 11].
Environmental factors are likely to contribute to IBD susceptibility; however, they are poorly understood and insufficiently described. The recent alarming rise in IBD incidence in the US points to changes in environmental factors rather than drastic alterations in allele frequency, since genetic remodeling could not occur over such a brief time . It seems plausible that environmental influences play as significant a role as their genetic counterparts and their combined actions dramatically affect disease presentation .
Study the incidence of SNP variants in NOD2, IL23r, OCTN1, and the IGR genes in a unique well-characterized rural Caucasian IBD population.
Examine haplotype frequencies in NOD2, IL23r, and IGR genes in both CD and UC patients.
Explore the SNPs' interaction, their possible effect on IBD susceptibility, and report a possible signature interaction model that could differentiate CD and UC patients.
Report risk factors that could serve as clinical predictors of IBD susceptibility
For the purposes of our genetic and clinical comparisons, we only examined the Caucasian subset in order to maintain statistical and descriptive homogeneity. Our patient population was composed of 1196 unrelated patients (63% women): 435 controls, 485 CD patients, and 276 UC patients.
Characteristics of CD patients using Vienna Classification
A1 (< 40 yrs)
A2 (> 40 yrs)
B1 (Non-stricturing, non-penetrating)
L1 (Terminal ileum)
L4 (Upper GI)
Analysis of Clinical Variables
Univariate analysis of UC vs. CD examined previously described relevant clinical factors associated with IBD. Gender, smoking status, number of IBD-associated operations, and family history of IBD were strong indicators of CD (p < 0.002). Age at diagnosis above 27 years old was more predictive of UC (p = 0.003).
SNP and Haplotype Analysis
Univariate Analysis of Allelic Association in SNPs
Single Nucleotide Polymorphism
Total # of Alleles (Total= 1644)
Control (Total= 582)
Crohn's Disease (Total= 654)
Ulcerative Colitis (Total= 408)
NOD2 - rs5743293
< 0.001 1
< 0.0001 2
< 0.001 4
NOD2 - rs2066844
< 0.001 1
< 0.001 2
NOD2 - rs2066845
< 0.001 1
< 0.001 2
IL-23r - rs11465804
NOD2 Haplotype Association Analysis
CD and UC combined
p- value (CD vs. UC)
rs574329 - rs2066844
rs574329 - rs2066845
rs574329 - rs2066844- rs2066845
IL23r Haplotype Association Analysis
CD and UC combined
(CD vs. UC)
rs10489629 - rs11465804
rs1004819 - rs10489629 -rs11465804
rs1004819 - rs2201841 - rs11465804 - rs11209026
rs1004819 - rs10489629 -rs11465804 - rs11209026
rs1004819 - rs10489629 - rs2201841 - rs11465804
rs1004819 - rs10489629 - rs2201841 - rs11465804 - rs11209026
Multivariate Analysis of Genetic and Clinical Predictors (CD vs. UC)
Multivariable Logistic Regression Analysis for UC vs. CD
Genetic and Clinical Variables
Ulcerative Colitis vs. Crohn's Disease
OR (95% CI)
Age at time of diagnosis
Number of Operations
Family History IBD
NOD2 - rs5743293
NOD2 - rs2066844
NOD2 - rs2066845
I. Current Diagnosis of IBD
IBD is a complex genetic disorder based on a multifaceted interaction between specific genes and environmental factors. The definitive diagnosis and correct identification of IBD remains difficult. IBD diagnostic markers may be lacking in the earliest stages of disease . Radiologic, endoscopic, and histological diagnostic approaches are commonly employed in identification of IBD; however, the inter-observer variability between pathologists calls into question their true clinical value. Distinction between CD and UC, particularly in the case of colonic IBD is important since surgical management varies in each disease, however, in many instances the pathological features overlap making differentiation problematic . Particularly in CD where variability of histological presentation makes accurate diagnosis a challenging exercise, pathologists often differ on their individual assessments . The apparent complexity of successful IBD diagnosis with current techniques demonstrates the immense value of developing a genetic and clinical screening tool.
II. Clinical and Environmental Predictors of IBD
There are several established clinical descriptors associated with IBD. Our analysis identified age at diagnosis, gender, smoking status, and family history of IBD as being the most important clinical predictors to distinguish UC from CD. Previous reports indicate that these factors are significantly related to IBD, but failed to adequately describe their effects.
A. Age at Time of Diagnosis
Loftus et al. described a systematic review of North Americans with CD and UC in respect to patient age at time of diagnosis. The mean age for CD diagnosis ranged from 33 - 45 years  while UC patients were diagnosed 5 - 10 years [18, 19] later. In concordance with these reports, the median age at time of diagnosis in our cohort was found to be 27 years. The age at time of diagnosis of ≥ 27 years age was found to be more predictive of UC patients (p = 0.04).
In a separate report, Loftus and Sandborn described the influence of gender on disease occurrence for CD and UC . They reported an increased incidence of UC among men, and CD among women . Our analysis concurs with this; the Caucasians in this study showed increased diagnosis of CD in women and UC in men (p = 0.014). The female predominance in CD could possibly be attributed to unspecified hormonal interactions .
C. Family History of IBD
The positive correlation between family history and IBD incidence seems plausible; this is possibly due to a combination of underlying genetic and environmental factors. This was shown in a study done at a University Hospital in Finland where a comparison among IBD patients revealed 16% of CD patients and 14% of UC patients had at least one family member afflicted with IBD [21, 22]. Our data strongly support these conclusions; i.e. a family history of IBD being more predictive of predisposition to CD than to UC (p < 0.0001).
D. Number of Surgeries
In addition to independent factors of gender, age, and family history, treatment approach may also be descriptive in IBD severity and progression. A North American population based study reported that 43% of patients with CD required major surgery involving incision, excision, and intestinal anastomosis compared to 40% for all UC hospitalizations . Our results showed similar findings; CD patients were more likely to require surgery than their UC counterparts (p = 0.015).
E. Smoking Status
Smoking was originally recognized as a risk factor for IBD susceptibility over 25 years ago [23, 24]. A meta-analysis performed by Mahid et al. showed that 12 of 13 reports found that "current" smoking status could be protective against UC development. In contrast, "former" smoker status was found to be associated with an increased risk for the development of UC. A parallel analysis of Crohn's patients revealed 6 of 9 reports indicating an increased disease risk associated with having "ever smoked" . Analysis of our patient cohort, "current" and "former" smoker statuses were both more associated with development of CD than UC (p < 0.001).
III. Genetic Predictors of IBD
NOD2 gene is identified to be in the IBD1 locus on chromosome 16 as a CD susceptibility gene. Hugot et al. mapped the IBD1 linkage locus and identified NOD2 as the underlying IBD susceptibility gene in 2001. Thirty non-conservative polymorphisms have been identified within this gene, but three single SNPs (rs2066844, rs2066845, and rs5743293) account for approximately 82% of the mutated alleles . Replication studies have confirmed that these three SNPs are independently associated with disease [11, 27, 28].
The receptor for the proinflammatory cytokine, IL-23r, also appears to play a key role in the progression of the chronic inflammation found in CD by driving inflammation through its role in the T-helper 17 (Th17) response. Activated myeloid and T cells express the IL-23 receptor and the level of its expression may affect the way Th17 cells create excess mucosal inflammation .
A. NOD2 and IL23r SNPs
In our patient population, all NOD2 variants and one IL23r variant were associated with IBD. After multivariable logistic regression analysis, only the NOD2 SNPs (rs2066844, rs2066845, and rs5743293) remained highly significant predictors of disease (p < 0.001), and in this context all were descriptive of CD, in agreement with previous reports. A meta-analysis conducted by Economou et al. found the same three NOD2 variants to be distinguishing risk factors for CD. According to their analyses, SNP rs5743293 carries a four-fold increase, rs2066845 a three-fold increase, and rs2066844 a two-fold increase of developing CD in non-Jewish descent Caucasians. They further reported that the combination of two of these NOD2 variants resulted in an increased odds of developing CD of 17 times the normal population . In contrast to NOD2 SNPs, we did not find the same genetic importance associated with IL23r polymorphisms in terms of describing and characterizing IBD. Although only one IL23r SNP was marginally significant in our univariate analysis, another variation of this allele yielded notable results when present with a NOD2 variant. A combination of NOD2 rs2066844 and IL23r rs2201841 was descriptive of patients with UC compared to controls (p = 0.04). No such association was found for CD patients.
B. Haplotype Analysis for NOD2 and IL23r
Another potential differentiator between CD and UC is haplotype analysis. Previous reports implicated numerous NOD2 and IL23r haplotype sequences associated to CD [30, 31]. The exact mechanism for disease pathogenesis in terms of these NOD2 and IL23r SNPs remains unclear, but an understanding of how allelic combinations alter the course of disease may prove useful. In our results, we found two haplotype sequences in NOD2 associated with CD: C-C-C and C-C-G (rs5743293-rs2066844-rs2066845). With respect to the IL23r gene, one haplotype was associated with CD and one with UC: Haplotypes C-A-C-G-G with CD and T-A-C-G-G with UC (rs1004819-rs10489629-rs2201841-rs11465804-rs11209026). Although potentially useful as genetic descriptors of CD and UC, the rarity of these haplotypes makes their clinical use unlikely.
Inflammatory bowel disease remains difficult to properly characterize, diagnose, and treat. Understanding the combined interactions between clinical, environmental, and genetic factors could serve as key in identifying IBD predictors as well as in properly differentiating between UC and CD. Our report implicates "female gender", "current" or "former" smoking status, positive family history of IBD, and NOD2 SNP minor alleles to be associated with CD. Clinically, this will help physicians identify patients with high risk for IBD development, differentiate CD from UC patients, and eventually help better tailor their medical/surgical treatment approach.
Population Selection and Classification
This study was approved by the University of Louisville Institutional Review Board (IRB). Written informed consent was obtained from all subjects. Genetic and clinical information was stored in a password-protected, prospectively maintained HIPAA compliant database. Patients were derived from a clinically well-described university-based IBD practice  and from a small relatively rural geographic area consisting of the state of Kentucky and southern Indiana. Our cohort is composed of 1337 patients (37% males and 63% females). It is comprised of unrelated individuals of various races (90% Caucasian, 7% African American, and 3% Asians). For the purpose of this study, we decided to study IBD in a uniform well-described group of patients, specifically Caucasians (n = 1196). These patients were not studied for IBD before. There is a profound environmental effect in our studied population as they are derived from one of the areas in the United States with the highest rates of adult cigarette smoking. Additionally, thirty seven percent of all IBD patients in our cohort have a positive family history of IBD, defined as a first or second degree relative with IBD, as compared to only 9% of controls. Perianal CD (PCD) occurred in 147 (46%) of CD patients. All patients were diagnosed to have IBD, using conventional radiologic, endoscopic, and/or histopathologic findings. In cases of IBD colitis, the diagnosis was confirmed by a single specialist gastrointestinal pathologist with a particular interest in IBD. CD patients and their clinical characteristics, in terms of age at diagnosis, disease location, and behavior were classified according to Vienna classification . Disease location in UC was classified as rectosigmoid, left-sided, or Pancolitis. Comparator patients to function as controls included patients from the same geographic area seen for non-inflammatory, non-neoplastic disorders including hemorrhoidal disease and screening colonoscopy.
Characterization of Environmental Risk
This included smoking status and it was verified by utilizing a validated Behavioral Risk Factor Surveillance Survey (BRFSS) that was prospectively administered.  "Current smoking" was defined as having smoked 100 cigarettes in one's lifetime and smoking every day or some days over the past six months.24 "Former smoking" was defined as having smoked 100 cigarettes in one's lifetime but currently no longer smoking at all.  A "non-smoker" is defined as someone who has not smoked 100 cigarettes in his or her lifetime and who does not currently smoke.24 The category "ever smoking" includes individuals from both the current and former smoking groups, while those who "never smoked" includes only non-smokers. 
Characterization of Clinical Data
Clinical data included gender, family history of IBD and cancer, age at time of diagnosis, presence of extra intestinal or perianal disease, number of IBD-associated surgeries (as a surrogate index for disease severity), and recurrence in CD.
Characterization of NOD2, IL23R, OCTN, and IGR Polymorphisms
Peripheral blood was obtained and genomic DNA was then extracted with a Puregene® DNA extraction kit (Gentra Systems Inc., Minneapolis, MN). Polymerase chain reaction (PCR) amplification and genotyping were performed on an ABI prism 7300 sequence detection system (Applied Biosystems©, Foster City, CA). TaqMan® SNP-specific PCR primers and fluorogenic probes were obtained from by Applied Biosystems (Applied Biosystems©, Foster City, CA). The fluorogenic minor groove binder (MGB) TaqMan® probes were labeled with a reporter dye, either FAM ([5-(&6)-carboxyfluorescein]) or VICs (a proprietary fluorescent dye produced by Applied Biosystems) specific for the wildtype and variant alleles of each of the following SNPs[33, 34] (additional file 1):
1- Three NOD2 SNPs: rs5743293, rs2066844, rs2066845.
2- Five IL23r SNPs: rs1004819, rs10489629, rs2201841, rs11465804, and rs11209026.
3- Two IGR SNPs: rs2522057 and rs7705189.
4- One OCTN SNP: rs1050152.
Descriptive statistics related to genetic and clinical characteristics were produced for the entire cohort. Categorical variables were compared using the Pearson Chi-square test (or Fisher's Exact test) for contingency tables . The t-test or Wilcoxon rank sum test was used to test the cohort for continuous variables. We also fit the univariable and multivariable logistic regression models for the probabilities of patients in "CD" or "UC" groups about their possible predictors . Furthermore, we examined haplotype associations between CD and UC.
We performed univariate analysis of our control population compared against CD and UC as well as CD against UC patients. Based on this analysis, we selected the following most significant IBD clinical predictors: age at time of diagnosis, gender, smoking status, IBD-associated surgery, and family history of IBD as descriptive of the IBD subtype. The age at time of diagnosis was dichotomized as < 27 or ≥ 27 years (as the median age at time of diagnosis was 27). Using these clinical criteria, we performed multivariable logistic regression analysis combining both significant genetic and clinical predictors.
We explored the genotype association, allelic association, and Hardy-Weinberg Equilibrium (HWE) test for 11 SNPs using the Pearson Chi-square test. For allelic association analysis of the 11 SNPs, we used the binary logistic regression method to estimate their odds ratios and 95% confidence intervals for patients with CD or UC as compared to control group. Odds ratios and 95% confidence intervals were estimated using binary logistic regression. Since all the studied genes had already been identified as associated genes for IBD, CD and UC susceptibility, no multiple testing corrections were required. Haplotype analysis was conducted for all the SNPs of NOD2, IL23r and IGR. All calculations were performed with Statistical Analysis Software (SAS) program [34, 36]. A p-value less than or equal to 0.05 was set to be significant.
This publication is made possible by the John and Caroline Price Family Trust and a gracious gift from Ms. Sara Shallenberger Brown. Additionally, this work was supported by NIH/NIEHS grant 1P30ES014443-01A1 to The Center for Environmental Genomics and Integrative Biology (CEGIB). These funding sources had no role in the production of this manuscript.
- Loftus EV: Clinical epidemiology of inflammatory bowel disease: Incidence, prevalence, and environmental influences. Gastroenterology. 2004, 126: 1504-1517. 10.1053/j.gastro.2004.01.063.View ArticlePubMed
- Martins NB, Peppercorn MA: Inflammatory bowel disease. Am J Manag Care. 2004, 10: 544-552.PubMed
- Mahid SS, Minor KS, Soto RE, Hornung CA, Galandiuk S: Smoking and inflammatory bowel disease: a meta-analysis. Mayo Clin Proc. 2006, 81: 1462-1471. 10.4065/81.11.1462.View ArticlePubMed
- Crawford NP, Colliver DW, Eichenberger MR, Funke AA, Kolodko V, Cobbs GA: CARD15 genotype-phenotype relationships in a small inflammatory bowel disease population with severe disease affection status. Dig Dis Sci. 2007, 52: 2716-2724. 10.1007/s10620-006-9208-z.View ArticlePubMed
- Mahid SS, Mulhall AM, Gholson RD, Eichenberger MR, Galandiuk S: Inflammatory bowel disease and African Americans: a systematic review. Inflamm Bowel Dis. 2008, 14: 960-967. 10.1002/ibd.20389.View ArticlePubMed
- Alvarez-Lobos M, Arostegui JI, Sans M, Tassies D, Plaza S, Delgado S: Crohn's disease patients carrying Nod2/CARD15 gene variants have an increased and early need for first surgery due to stricturing disease and higher rate of surgical recurrence. Ann Surg. 2005, 242: 693-700. 10.1097/01.sla.0000186173.14696.ea.PubMed CentralView ArticlePubMed
- Armuzzi A, Ahmad T, Ling KL, de SA, Cullen S, van HD: Genotype-phenotype analysis of the Crohn's disease susceptibility haplotype on chromosome 5q31. Gut. 2003, 52: 1133-1139. 10.1136/gut.52.8.1133.PubMed CentralView ArticlePubMed
- Babusukumar U, Wang T, McGuire E, Broeckel U, Kugathasan S: Contribution of OCTN variants within the IBD5 locus to pediatric onset Crohn's disease. Am J Gastroenterol. 2006, 101: 1354-1361. 10.1111/j.1572-0241.2006.00564.x.View ArticlePubMed
- Lin Z, Poritz L, Franke A, Li TY, Ruether A, Byrnes KA: Genetic association of nonsynonymous variants of the IL23R with familial and sporadic inflammatory bowel disease in women. Dig Dis Sci. 2010, 55: 739-746. 10.1007/s10620-009-0782-8.View ArticlePubMed
- Inohara , Chamaillard , McDonald C, Nunez G: NOD-LRR proteins: role in host-microbial interactions and inflammatory disease. Annu Rev Biochem. 2005, 74: 355-383. 10.1146/annurev.biochem.74.082803.133347.View ArticlePubMed
- Hampe J, Grebe J, Nikolaus S, Solberg C, Croucher PJ, Mascheretti S: Association of NOD2 (CARD 15) genotype with clinical course of Crohn's disease: a cohort study. Lancet. 2002, 359: 1661-1665. 10.1016/S0140-6736(02)08590-2.View ArticlePubMed
- Danese S, Sans M, Fiocchi C: Inflammatory bowel disease: the role of environmental factors. Autoimmun Rev. 2004, 3: 394-400. 10.1016/j.autrev.2004.03.002.View ArticlePubMed
- Fiocchi C: Inflammatory bowel disease: etiology and pathogenesis. Gastroenterology. 1998, 115: 182-205. 10.1016/S0016-5085(98)70381-6.View ArticlePubMed
- Geboes K: Pathology of inflammatory bowel diseases (IBD): variability with time and treatment. Colorectal Dis. 2001, 3: 2-12.View ArticlePubMed
- Yantiss RK, Odze RD: Diagnostic difficulties in inflammatory bowel disease pathology. Histopathology. 2006, 48: 116-132. 10.1111/j.1365-2559.2005.02248.x.View ArticlePubMed
- Lessells AM, Beck JS, Burnett RA, Howatson SR, Lee FD, McLaren KM: Observer variability in the histopathological reporting of abnormal rectal biopsy specimens. J Clin Pathol. 1994, 47: 48-52. 10.1136/jcp.47.1.48.PubMed CentralView ArticlePubMed
- Loftus EV, Schoenfeld P, Sandborn WJ: The epidemiology and natural history of Crohn's disease in population-based patient cohorts from North America: a systematic review. Aliment Pharmacol Ther. 2002, 16: 51-60. 10.1046/j.1365-2036.2002.01140.x.View ArticlePubMed
- Bjornsson S, Johannsson JH: Inflammatory bowel disease in Iceland, 1990-1994: a prospective, nationwide, epidemiological study. Eur J Gastroenterol Hepatol. 2000, 12: 31-38. 10.1097/00042737-200012010-00007.View ArticlePubMed
- Loftus EV, Silverstein MD, Sandborn WJ, Tremaine WJ, Harmsen WS, Zinsmeister AR: Ulcerative colitis in Olmsted County, Minnesota, 1940-1993: incidence, prevalence, and survival. Gut. 2000, 46: 336-343. 10.1136/gut.46.3.336.PubMed CentralView ArticlePubMed
- Loftus EV, Sandborn WJ: Epidemiology of inflammatory bowel disease. Gastroenterol Clin North Am. 2002, 31: 1-20. 10.1016/S0889-8553(01)00002-4.View ArticlePubMed
- Bernstein CN, Nabalamba A: Hospitalization, surgery, and readmission rates of IBD in Canada: a population-based study. Am J Gastroenterol. 2006, 101: 110-118. 10.1111/j.1572-0241.2006.00330.x.View ArticlePubMed
- Halme L, Turunen U, Helio T, Paavola P, Walle T, Miettinen A: Familial and sporadic inflammatory bowel disease: comparison of clinical features and serological markers in a genetically homogeneous population. Scand J Gastroenterol. 2002, 37: 692-698. 10.1080/00365520212511.View ArticlePubMed
- Harries AD, Baird A, Rhodes J: Non-smoking: a feature of ulcerative colitis. Br Med J (Clin Res Ed). 1982, 284: 706-10.1136/bmj.284.6317.706.View Article
- Somerville KW, Logan RF, Edmond M, Langman MJ: Smoking and Crohn's disease. Br Med J (Clin Res Ed). 1984, 289: 954-956. 10.1136/bmj.289.6450.954.View Article
- Hugot JP, Chamaillard M, Zouali H, Lesage S, Cezard JP, Belaiche J: Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn's disease. Nature. 2001, 411: 599-603. 10.1038/35079107.View ArticlePubMed
- Lesage S, Zouali H, Cezard JP, Colombel JF, Belaiche J, Almer S: CARD15/NOD2 mutational analysis and genotype-phenotype correlation in 612 patients with inflammatory bowel disease. Am J Hum Genet. 2002, 70: 845-857. 10.1086/339432.PubMed CentralView ArticlePubMed
- Cuthbert AP, Fisher SA, Mirza MM, King K, Hampe J, Croucher PJ: The contribution of NOD2 gene mutations to the risk and site of disease in inflammatory bowel disease. Gastroenterology. 2002, 122: 867-874. 10.1053/gast.2002.32415.View ArticlePubMed
- Vermeire S, Wild G, Kocher K, Cousineau J, Dufresne L, Bitton A: CARD15 genetic variation in a Quebec population: prevalence, genotype-phenotype relationship, and haplotype structure. Am J Hum Genet. 2002, 71: 74-83. 10.1086/341124.PubMed CentralView ArticlePubMed
- Economou M, Trikalinos TA, Loizou KT, Tsianos EV, Ioannidis JP: Differential effects of NOD2 variants on Crohn's disease risk and phenotype in diverse populations: a metaanalysis. Am J Gastroenterol. 2004, 99: 2393-2404. 10.1111/j.1572-0241.2004.40304.x.View ArticlePubMed
- Okazaki T, Wang MH, Rawsthorne P, Sargent M, Datta LW, Shugart YY: Contributions of IBD5, IL23R, ATG16L1, and NOD2 to Crohn's disease risk in a population-based case-control study: evidence of gene-gene interactions. Inflamm Bowel Dis. 2008, 14: 1528-1541. 10.1002/ibd.20512.PubMed CentralView ArticlePubMed
- Sugimura K, Taylor KD, Lin YC, Hang T, Wang D, Tang YM: A novel NOD2/CARD15 haplotype conferring risk for Crohn disease in Ashkenazi Jews. Am J Hum Genet. 2003, 72: 509-518. 10.1086/367848.PubMed CentralView ArticlePubMed
- Gasche C, Scholmerich J, Brynskov J, D'Haens G, Hanauer SB, Irvine EJ: A simple classification of Crohn's disease: report of the Working Party for the World Congresses of Gastroenterology, Vienna 1998. Inflamm Bowel Dis. 2000, 6: 8-15.View ArticlePubMed
- Applied Biosystems. Accessed July, 2010, http://www.appliedbiosystems.com
- Fleiss JL, Levin BA, Paik MC: Statistical methods for rates and proportions. 2003, Hoboken, N.J: Wiley-Interscience, 3View Article
- Agresti A: Categorical data analysis. 2002, New York: Wiley, 2View Article
- The SAS System V9. 2003, Cary, NC, SAS Institute Inc
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.