Gastroenterology
Volume 130, Issue 3 , Pages 650-656, March 2006

Predictors of Crohn’s Disease

Department of Gastroenterology, Saint-Antoine Hospital, and Pierre et Marie Curie University, Paris, France

Received 27 July 2005; accepted 16 November 2005. published online 13 December 2005.

Article Outline

Background & Aims: Early intensive therapy in Crohn’s disease should be considered only in patients with disabling disease. The aim of our study was to identify at diagnosis factors predictive of a subsequent 5-year disabling course. Methods: Among the 1526 patients seen at our unit with Crohn’s disease diagnosed between 1985 and 1998, we excluded patients operated on within the first month of the disease, patients with inadequate data, and patients with severe chronic nondigestive disease. In the 1188 remaining patients, Crohn’s disease course within the first 5 years of the disease was categorized as disabling when at least 1 of the criteria of clinical severity, conventionally predefined, was present. Results: Among the 1123 patients with follow-up data allowing full 5-year course classification, the rate of disabling disease was 85.2%. Independent factors present at diagnosis and significantly associated with subsequent 5-year disabling were the initial requirement for steroid use (OR 3.1 [95% CI: 2.2–4.4]), an age below 40 years (OR 2.1 [95% CI: 1.3–3.6]), and the presence of perianal disease (OR 1.8 [95% CI: 1.2–2.8]). The positive predictive value of disabling disease in patients with 2 and 3 predictive factors of disabling disease was 0.91 and 0.93, respectively. These values were 0.84 and 0.91, respectively, when tested prospectively in an independent group of 302 consecutive patients seen at our institution from 1998. Conclusions: At diagnosis of Crohn’s disease in a referral center, factors predictive of subsequent 5-year disabling course are an age below 40 years, the presence of perianal disease, and the initial requirement for steroids.

Abbreviation used in this paper:  CD, Crohn’s disease

 

Crohn’s disease (CD) is a chronic idiopathic inflammatory condition of the gastrointestinal tract that is often disabling through chronic symptomatic inflammation and/or frequent clinical relapses and/or clinical consequences of intestinal irreversible lesions or intestinal resections. During the last decades, significant advances in the therapeutic potential of CD have been made. The overall efficacy and good tolerance of immunosuppressants have been demonstrated, leading to their larger use, and the first potent biologic therapy, ie, infliximab, has become available. In parallel, the global therapeutic strategy for CD has been modelized according to 2 options. The step-up strategy consists in adapting individually, on a sequential basis, the level of antiinflammatory therapy to the anatomoclinical severity of disease. The top-down strategy is based on the very early use of intensive therapy (immunosuppressants and/or biologics) to maintain a good quality of life from the first flare-up of the disease and prevent any irreversible consequence of the disease. The step-up option is currently the standard approach, but recent data suggest that this strategy has not clearly changed the global outcome of CD1 nor significantly reduced the need for surgery.2, 3 Thus, it is now crucial to evaluate the top-down strategy4 and to address 2 major points in this field. First, benefits of the top-down strategy should be shown as outweighing risks, such as lymphomas for immunosuppressants5, 6, 7 and severe infections for biotechnologic treatments.8 Second, a particular effort should be made to identify patients with spontaneous disabling course of the disease, who should be properly considered for enrollment in early intensive therapy trials. Two concordant works on natural history of CD in the early 1990s suggested that, in an individual, the severity of the disease within the first 2 to 3 years of the disease tends to remain unchanged thereafter.9, 10 However, these works and following studies on natural history of CD failed to identify simple clinical criteria present at diagnosis of the disease and predictive of its further evolution. The aim of our work was, from a large retrospective single cohort of patients, to identify at diagnosis factors predictive of a subsequent 5-year disabling course and to test prospectively the accuracy of these factors in an independent test group seen at the same referral center.

Back to Article Outline

Patients and Methods 

Patient Selection 

We considered for inclusion in the retrospective part of the study all the patients with CD seen consecutively at our unit between January 1985 and January 1998. Diagnosis of CD was based on Lennard-Jones criteria.11 Among the 1526 patients screened, we excluded patients operated on at diagnosis (n = 62), patients with incomplete medical data regarding either the diagnosis period or the time interval between diagnosis and the end of the 5-year study period or loss to follow-up (n = 263), and patients with severe chronic nondigestive disease (n = 13). Thus, 1188 patients were included in the study.

Study Design 

We arbitrarily defined CD course as disabling during the 5-year period following diagnosis when at least 1 of the following criteria was present in this time interval: more than 2 steroid courses required and/or dependence on steroids12; further hospitalization after diagnosis for flare-up or complication of the disease; presence of disabling chronic symptoms (cumulative time of more than 12 months of disabling symptoms (diarrhea with nocturnal and/or urgent stools, intense abdominal pain because of intestinal obstruction, fever, fatigue attributable to the disease, joint pain, painful uveitis or pyoderma gangrenosum) within the 5-year study period); need for immunosuppressive therapy; and intestinal resection or surgical operation for perianal disease. According to these criteria, the CD clinical course of an individual patient with incomplete 5-year follow-up after diagnosis could be classified adequately as disabling if at least 1 criterion of disabling disease (for instance hospitalization for a second flare-up) was observed in the time interval between diagnosis and the end of the follow-up. Conversely, to be adequately classified as having a nondisabling disease in the 5-year period following diagnosis, patients needed to have a complete 5-year follow-up after diagnosis and no criterion of disabling disease during this interval. According to these classification rules, the individual course of the 1188 patients included in the study was retrospectively abstracted from medical files by 2 senior gastroenterologists (L.B. and J.C.).

To estimate the prevalence of patients with disabling disease, we restricted data analysis to the subset of 262 patients who were managed at our institution as outpatients or inpatients during the period of the first flare-up of the disease. The justification of this choice was that patients referred to our tertiary center after the first flare-up of the disease were expected to have a more severe disease than population-based patients.

To determine the factors predictive of CD disabling course within the 5-year period following diagnosis (see the statistics section), we restricted the analysis to the subset of patients with complete follow-up during the 5-year period following diagnosis and patients with a follow-up time inferior to 5 years but exhibiting at least 1 criterion of disabling disease during the time interval between diagnosis and loss of follow-up. Among these patients, search for factors predictive of disabling disease was performed by univariate tests followed by multivariate analysis.

In the last part of the study, we tested prospectively the positive predictive value of factors predictive of 5-year disabling disease derived from the multivariate analysis. This part of the study was performed in the first 302 consecutive patients seen at our institution from 1998 with expected time of medical follow-up above 5 years after diagnosis. Review of medical data was performed according to the same method and the same classification rules as in the patients seen between 1985 and 1998. Again, the CD clinical course of an individual patient with incomplete 5-year follow-up after diagnosis could be classified adequately as disabling if at least 1 criterion for disabling disease (for instance hospitalization for second flare-up) was observed in the time interval between diagnosis and the end of the follow-up. Conversely, to be adequately classified as having a nondisabling disease in the 5-year period following diagnosis, patients needed to have a complete 5-year follow-up and no criterion of disabling disease during this interval.

Data Analysis 

Among the 262 patients managed at our institution from the first flare-up of CD, we estimated the prevalence of disabling disease according to 2 extreme hypotheses. The first hypothesis was that all the patients lost for follow-up before the end of the 5-year study period and classified as having a nondisabling disease for the time interval between diagnosis and loss of follow-up had a disabling disease for the whole 5-year period after diagnosis. That means that all these patients were supposed to have experienced an event marker of disabling disease in the time interval between loss to follow-up and the end of the 5-year period after diagnosis. This yielded a first estimated prevalence of patients with disabling CD, with its corresponding 95% confidence interval (CI). The second hypothesis was that all the patients lost for follow-up before the end of the 5-year study period and classified as having a nondisabling disease for the time interval between diagnosis and loss of follow-up remained with nondisabling disease for the whole 5-year period after diagnosis. That means that all these patients were supposed to have experienced no event marker of disabling disease in the time interval between loss of follow-up and the end of the 5-year period after diagnosis. This yielded a second estimated prevalence of patients with disabling CD, with its corresponding 95% CI. From the 2 overlapping confidence intervals, we proposed a resulting confidence interval with the lowest lower limit of the 2 confidence intervals as lower limit and the highest upper limit of the 2 confidence intervals as upper limit.

Search for factors available at diagnosis and predictive of subsequent 5-year disabling CD course was performed in the subset of patients with complete follow-up during the 5-year period following diagnosis and patients with a follow-up time inferior to 5 years but exhibiting at least 1 criterion of disabling disease during the time interval between diagnosis and loss of follow-up. Calculations were performed using Statview statistical software (version 5.0, SAS Institute Inc., Cary, NC). Nine variables were tested by univariate analysis: sex; ethnic origin; age at diagnosis (categorized according Vienna classification: below 40 years, equal to or above 40 years)13; location of the disease at diagnosis (small bowel only, small bowel and colon, colon only); prior appendectomy; smoking status (current smoker [>1 cigarette/day] vs never or exsmoker); systemic manifestations at diagnosis; perianal disease at diagnosis; and steroids required for the treatment of the first flare. Variables were compared using the χ2 test or the Fisher exact test when appropriate. Those variables with P values below .10 were further tested in a logistic multivariate regression model using a backward stepwise procedure. The final step retained independent factors with P values below .05 in a 2-tailed test. The positive predictive value of the factors present at diagnosis and independently associated with subsequent 5-year disabling disease was calculated for each value of a score (the disabling CD predictive score) defined as the arithmetical sum of the factors present in an individual patient. The positive predictive values (of subsequent 5-year disabling disease) of the different values of the score were first calculated in the original group of patients who contributed to the statistical elaboration of the predictive factors. In a second step, the performances of the score were assessed according to the same method in a distinct subsequent group of 302 patients seen at our institution after 1998 (see the preceding Methods section). The 5-year probability of complications in the retrospective group was calculated using the log-rank method.

Back to Article Outline

Results 

Patient Populations 

Among the 262 patients managed at our institution from the first flare-up of the disease, 105 (40.0%) were males and 157 (60.0%) were females, with a median age at diagnosis of 27.8 years (range, 12.6–83.2 years). CD initially involved small bowel only, small bowel and colon, and colon alone in 26.0%, 37.8%, and 36.2% of the patients, respectively. Perianal lesions were present at diagnosis in 62 (23.7%) patients. Steroids were required for treating the first flare-up in 170 (64.9%) patients.

Search for factors available at diagnosis predictive of subsequent 5-year disabling CD course was performed in the 1123 patients with complete follow-up during the 5-year period following diagnosis and patients with a follow-up time inferior to 5 years but exhibiting at least 1 criterion on disabling disease during the time interval between diagnosis and loss of follow-up. Among these patients, 425 (37.9%) were males and 698 (62.1%) were females, with a median age at diagnosis of 25.3 years (range, 6.7–87.4 years). CD initially involved small bowel only, small bowel and colon, or colon alone in 34.5%, 37.4%, and 28.1% of the patients, respectively. Perianal lesions were present at diagnosis in 282 (25.1%) patients. Steroids were required for treating the first flare-up in 665 (59.2%) patients.

Prevalence of Disabling CD in the 5-Year Period Following Diagnosis 

Among the 262 patients managed at our institution from the first flare-up of the disease, 41 were lost for follow-up before the end of the 5-year study period and classified as having a nondisabling disease for the time interval between diagnosis and loss of follow-up (Figure 1). Based on the hypothesis that all the 41 patients had a nondisabling disease for the whole 5-year period after diagnosis, the estimated prevalence of disabling disease was 64.9% (95% CI: 59.1%–70.7%). Based on the hypothesis that all the 41 patients had a disabling disease for the whole 5-year period after diagnosis, the estimated prevalence of disabling disease was 80.5% (95% CI: 68.8%–92.2%). The resulting global 95% CI of the estimated prevalence of disabling disease was 59.1% to 92.2%.

  • View full-size image.
  • Figure 1. 

    Flow chart describing patient selection for analysis, starting from the 1526 patients seen consecutively at our unit from January 1985 to January 1998. Pts, patients; CD, Crohn’s disease; FU, follow-up.

Factors Predictive of Disabling CD Course During the 5-Year Period Following Diagnosis 

A total of 1123 patients with follow-up data allowing complete classification (957 [85.2%] patients with disabling disease and 166 [14.8%] patients with nondisabling disease during the 5-year period after diagnosis) was analyzed. By univariate analysis, 3 factors present at diagnosis were significantly associated with a disabling 5-year course after diagnosis (Table 1). These factors were an age below 40 years (P = .0004), the initial localization of the disease restricted to the colon (P = .002), the presence of perianal lesions (P = .01), and a steroid requirement for treating the first flare-up of the disease (P = .0001). The 5-year probability of disease complications between the 2 groups is given in Table 2. By logistic regression, 3 factors were independently predictive of a disabling CD course in the 5-year period following diagnosis: the initial requirement for steroids (OR: 3.1 [95% CI: 2.2–4.4]), an age at diagnosis below 40 years (OR: 2.1 [95% CI: 1.3–3.6]), and the presence of perianal disease at diagnosis (OR: 1.8 [95% CI: 1.2–2.8]).

Table 1. Distribution of Patient Characteristics at Diagnosis According to the Nondisabling or Disabling Clinical Course in the 5-Year Period After Diagnosis in the 1123 Patients With Follow-up Data Allowing Full 5-Year Classification
Variables5-Year clinical course after diagnosisStatistical univariate testa
Nondisabling, % (n = 166)Disabling, % (n = 957)
Sex
Male40.437.3NS
Female59.662.7
Ethnic origin
White91.086.9NS
Nonwhite9.013.1
Age at onset
Below 40 y77.187.7P=.0004
40 y or above22.912.3
Location of the disease
Small bowel only44.632.8P=.002
Small bowel and colon25.939.4
Colon only29.527.8
Previous appendectomy
Yes30.631.1NS
No69.468.9
Smoking status
Smoker50.357.4P=.09
Ex- or nonsmoker49.742.6
Systemic manifestations
Yes44.648.6NS
No55.451.4
Perianal lesions at diagnosis
Yes17.526.4P=.01
No82.573.6
Requirement for steroids for treating the first flare
Yes37.365.2P=.0001
No62.734.8

NS, not significant.

a P values are given for P < .10.

Table 2. Five-Year Probability of Disease Complications According to the Disabling or Nondisabling Clinical Course in the 5-Year Period After Diagnosis in the 1123 Patients With Follow-up Data Allowing Full 5-Year Classification
Patients with nondisabling disease (n = 166)Patients with disabling disease (n = 957)
Treatment by immunosuppressants0.000.43
Intestinal resection0.000.35
Definite stoma0.000.02
Intestinal fistula or abscess0.000.19
Stricture formation0.01a0.18

a Asymptomatic strictures or strictures not associated with chronic disabling symptoms.

Given the results of the multivariate analysis, the values of the disabling CD course predictive score ranged from 0 (none of the 3 independent factors associated with a disabling subsequent course) to 3 (all 3 factors present in the same individual). Regarding the risk of developing a disabling CD course in the 5-year period following diagnosis, the positive predictive value of the disabling CD predictive score in an individual patient ranged from 0.61 to 0.93 (Figure 2). Positive predictive value was 0.91 and 0.93 in patients having a score equal to 2 and 3, respectively.

  • View full-size image.
  • Figure 2. 

    Distribution of the patients and predictive positive value of having a disabling Crohn’s disease course in the 5-year period after diagnosis according to the value of the disabling Crohn’s disease predictive score.

The score was secondarily tested in an independent sample of 302 consecutive patients who were seen at our institution after 1998 and who had follow-up data allowing complete classification. Among these patients, 113 (37.4%) were males and 189 (62.6%) were females, with a median age at diagnosis of 23.2 years (range, 8.5–76.9 years). CD initially involved small bowel only, small bowel and colon, or colon alone in 37.8%, 33.4%, and 28.8% of the patients, respectively. Perianal lesions were present at diagnosis in 72 (23.8%) patients. Steroids were required for treating the first flare-up in 180 (59.6%) patients. In this prospective group of patients, positive predictive value of the CD course predictive score was 0.84 and 0.91 in patients having a score equal to 2 and 3, respectively.

Back to Article Outline

Discussion 

Our results suggest that most patients with CD seen at referral centers will experience a disabling clinical course in the 5-year period following diagnosis. We identified at diagnosis 3 factors (steroids required for treating the first flare-up, age below 40 years, and presence of perianal disease at diagnosis) independently associated with an increased risk of subsequent 5-year disabling CD clinical course. In our cohort, the presence at diagnosis in an individual patient of 2 or 3 out of the 3 factors associated with disabling course was well predictive of the subsequent 5-year evolution, both in the founding study and in a prospective validation test in an independent patient sample.

The concept of nondisabling CD clinical course is evolving in the mind of both physicians and patients. The therapeutic potential of biologic therapy,14 in addition to the established 5-year efficacy of immunosuppressants,15 makes it possible to consider in the future an optimal therapeutic response in CD as a sustained healing of intestinal lesions in parallel with prolonged clinical remission.16, 17 According to this evolving view of therapeutic goals, a nondisabling disease should be considered progressively as very close to normal life after diagnosis. Accordingly, the quality of life of patients with nondisabling disease should be most of the time identical to that of the general population. At this time and in our study, we chose arbitrarily as markers of disabling disease all the events that can interfere with the social, professional, and private life of the patients in the 5-year period after diagnosis. These events included of course surgical treatment of CD lesions and hospitalizations for CD flare-ups that are now taken into account when comparing treatment strategies.18 Given, first, that numerous adverse effects of steroids can affect the patients psychologically and, second, that steroids are required for treating moderate to severe flares, we considered as disabling the CD course of patients experiencing more than 2 flare-ups requiring the use of steroids in the 5-year period after diagnosis. Last, we considered as markers of chronic active disease (therefore disabling) the presence of disabling chronic symptoms and the need for immunosuppressive therapy. Despite the increasing and earlier use of immunosuppressants in our institution,19 our global therapeutic policy still consists in the step-up approach. In this strategy, the first use of immunosuppressants is considered only in patients with disabling disease15, 20, 21, 22 and remains a good severity marker.23

In our cohort, most of the patients were classified as having a disabling CD course during the 5-year period after diagnosis. This proportion held true when we restricted the analysis to patients managed at our institution from the first flare-up of the disease, considering that patients referred to our tertiary center after the first flare-up of the disease are expected to have a more severe disease than population-based patients.24 Although the restricted group of patients seen by us at diagnosis was not population-based, the median age at diagnosis in this group was similar to that of huge population-based cohorts,25 despite the fact that it has been recently suggested that patients with late onset are underrepresented at tertiary referral centers.24 In addition, the percentage of patients treated by steroids for the first flare-up (65%) was close to the percentage of patients who received steroids within the first year of disease in a North European population-based study.12 Based on these similarities, we suggest that the majority of patients of the CD general population are prone to have a disabling disease course in the 5-year period after diagnosis. However, referral bias may still hold in our population so that population-based studies are needed before generalization of our assumption.

As the use of immunomodulators increases with time, one could suspect that this continuous trend introduced a bias in our classification. In our center, the global treatment policy is still a step-up one so that the initiation of immunomodulators is decided in the case of disabling symptoms and is poorly prone to affect per se the classification. When dividing our retrospective population in 4 quartiles according to the date of diagnosis between 1985 and 1998, the rate of patients with disabling disease and the 5-year probability of intestinal resection did not change significantly over time, whereas the 5-year probability of receiving immunomodulators progressively increased (data not shown). These findings are in accordance with our published experience showing that the operative risk did not significantly change over time in our center, despite the increased and earlier use of immunosuppressants.3

We identified 3 clinical features present at diagnosis and independently associated with a subsequent disabling disease. By contrast, in the 2 previous concordant works that evidenced that the individual course pattern of a patient tends to remain the same throughout the first years of the disease, no factor available at diagnosis was shown to correlate with the subsequent course of the disease.9, 10 However, in these studies, the total number of patients included was lower than in our cohort, and no multivariate analysis was performed. In addition, in 1 of the studies, there was a trend toward a later age at onset and a lower prevalence of anal lesions at diagnosis in the group of patients with nondisabling disease in the 3-year period following diagnosis.9

The standard global therapeutic approach currently recommended in Crohn’s disease is the step-up strategy.15, 17, 20 However, controlled trials have been designed for evaluating the benefit/risk ratio of the top-down strategy, consisting of the use of purine analogues and/or biotherapies from the first flare-up of the disease. Regarding purine analogues, Markowitz et al have shown in children requiring steroids for the treatment of the first flare-up of CD that the very early use of 6-mercaptopurine was associated with steroid sparing and a more favorable clinical outcome in the 18-month period following diagnosis.26 Regarding the potential clinical benefit of a very early use of monoclonal antibodies to tumor necrosis factor, one must keep in mind the safety profile of these compounds, with a possibly attributable increased risk of death up to 1%,8 essentially related to the increased risk of serious infections.27, 28 In this context, it is questionable from an ethical point of view to include in trials matching classic treatment vs the very early use of monoclonal antibodies to tumor necrosis factor patients who would have experienced a spontaneous 5-year nondisabling disease. In our study, we have shown that the presence in an individual patient of 2 or 3 out of the 3 factors associated with disabling course (that is a value of the disabling CD predictive score of 2 or 3) is predictive of a disabling CD course in the subsequent 5-year period, with a confidence level of approximately 90%. This score should be validated or invalidated rapidly in other tertiary or population-based cohorts. If cross validation is achieved, the score should be considered as an eligibility criterion in trials including arms with aggressive early therapy.

Back to Article Outline

References 

  1. Wolters FL , Russel MG , Stockbrugger RW . Systematic review (has disease outcome in Crohn’s disease changed during the last four decades?) . Aliment Pharmacol Ther . 2004;20:483–496
  2. Sands BE, Arsenault JE, Rosen MJ, Alsahli M, Bailen L, Banks P, et al. Risk of early surgery for Crohn’s disease (implications for early treatment strategies) . Am J Gastroenterol . 2003;98:2712–2718
  3. Cosnes J , Nion-Larmurier I , Beaugerie L , Afchain P , Tiret E , Gendre JP . Impact of the increasing use of immunosuppressants in Crohn’s disease on the need for intestinal surgery . Gut . 2005;54:237–241
  4. Hommes D, Baert F, Van Assche G, Caenepeel F, Vergauwe P, Tuynman H, et al. Management of recent onset Crohn’s disease (a controlled, randomized trial comparing step-up and top-down therapy) . Gastroenterology . 2005;129:371
  5. Dayharsh GA, Loftus EV, Sandborn WJ, Tremaine WJ, Zinsmeister AR, Witzig TE, et al. Epstein-Barr virus-positive lymphoma in patients with inflammatory bowel disease treated with azathioprine or 6-mercaptopurine . Gastroenterology . 2002;122:72–77
  6. Lewis JD , Bilker WB , Brensinger C , Deren JJ , Vaughn DJ , Strom BL . Inflammatory bowel disease is not associated with an increased risk of lymphoma . Gastroenterology . 2001;121:1080–1087
  7. Kandiel A , Fraser AG , Korelitz BI , Brensinger C , Lewis JD . Increased risk of lymphoma among inflammatory bowel disease patients treated with azathioprine and 6-mercaptopurine . Gut . 2005;54:1121–1125
  8. Colombel JF, Loftus EV, Tremaine WJ, Egan LJ, Harmsen WS, Schleck CD, et al. The safety profile of infliximab in patients with Crohn’s disease (the Mayo clinic experience in 500 patients) . Gastroenterology . 2004;126:19–31
  9. Beaugerie L, Le Quintrec Y, Paris JC, Godchaux JM, Saint-Raymond A, Schmitz J, et al. Testing for course patterns in Crohn’s disease using clustering analysis . Gastroenterol Clin Biol . 1989;13:1036–1041
  10. Munkholm P , Langholz E , Davidsen M , Binder V . Disease activity courses in a regional cohort of Crohn’s disease patients . Scand J Gastroenterol . 1995;30:699–706
  11. Lennard-Jones JE . Classification of inflammatory bowel disease . Scand J Gastroenterol Suppl . 1989;170:2–6 discussion 16–19
  12. Munkholm P , Langholz E , Davidsen M , Binder V . Frequency of glucocorticoid resistance and dependency in Crohn’s disease . Gut . 1994;35:360–362
  13. Gasche C, Scholmerich J, Brynskov J, D’Haens G, Hanauer SB, Irvine EJ, et al. 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
  14. Sandborn WJ , Targan SR . Biologic therapy of inflammatory bowel disease . Gastroenterology . 2002;122:1592–1608
  15. Feagan BG . Maintenance therapy for inflammatory bowel disease . Am J Gastroenterol . 2003;98:S6–S17
  16. Rutgeerts P . A critical assessment of new therapies in inflammatory bowel disease . J Gastroenterol Hepatol . 2002;17(Suppl):S176–S185
  17. Hanauer SB . Crohn’s disease (step up or top down therapy) . Best Pract Res Clin Gastroenterol . 2003;17:131–137
  18. Rutgeerts P, Feagan BG, Lichtenstein GR, Mayer LF, Schreiber S, Colombel JF, et al. Comparison of scheduled and episodic treatment strategies of infliximab in Crohn’s disease . Gastroenterology . 2004;126:402–413
  19. Cosnes J , Nion-Larmurier I , Beaugerie L , Gendre JP . Do immunosuppressants have any impact on the need for surgery in Crohn’s disease? . Gastroenterology . 2003;124:A10
  20. Sandborn WJ . Evidence-based treatment algorithm for mild to moderate Crohn’s disease . Am J Gastroenterol . 2003;98:S1–S5
  21. Hanauer SB , Present DH . The state of the art in the management of inflammatory bowel disease . Rev Gastroenterol Disord . 2003;3:81–92
  22. Carter MJ , Lobo AJ , Travis SP . Guidelines for the management of inflammatory bowel disease in adults . Gut . 2004;53(Suppl 5):V1–V16
  23. Cosnes J , Beaugerie L , Carbonnel F , Gendre JP . Smoking cessation and the course of Crohn’s disease (an intervention study) . Gastroenterology . 2001;120:1093–1099
  24. Zankel E , Rogler G , Andus T , Reng CM , Scholmerich J , Timmer A . Crohn’s disease patient characteristics in a tertiary referral center (comparison with patients from a population-based cohort) . Eur J Gastroenterol Hepatol . 2005;17:395–401
  25. Lapidus A , Bernell O , Hellers G , Persson PG , Lofberg R . Incidence of Crohn’s disease in Stockholm County 1955–1989 . Gut . 1997;41:480–486
  26. Markowitz J , Grancher K , Kohn N , Lesser M , Daum F . A multicenter trial of 6-mercaptopurine and prednisone in children with newly diagnosed Crohn’s disease . Gastroenterology . 2000;119:895–902
  27. Keane J, Gershon S, Wise RP, Mirabile-Levens E, Kasznica J, Schwieterman WD, et al. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent . N Engl J Med . 2001;345:1098–1104
  28. Warris A , Bjorneklett A , Gaustad P . Invasive pulmonary aspergillosis associated with infliximab therapy . N Engl J Med . 2001;344:1099–1100

PII: S0016-5085(05)02529-1

doi:10.1053/j.gastro.2005.12.019

Refers to article:

  • Continuing Medical Education Exams 3: March 2006: The Aer-O-Scope: Proof of Concept of a Pneumatic, Skill Independent, Self Propelling, Self Navigating Colonoscope

    B. Vucelic, D. Rex, R. Pulanic, J. Pfefer, I. Hrstic, B. Levin, Z. Halpren, N. Arber
    Gastroenterology March 2006 (Vol. 130, Issue 3, Page 930)

Gastroenterology
Volume 130, Issue 3 , Pages 650-656, March 2006

Article Tools

* Image Usage & Resolution