SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MUCOSAL MARKERS OF INFLAMMATION PREDICTIVE OF RELAPSE
  5. INTESTINAL PERMEABILITY AS PREDICTOR
  6. SYSTEMIC INFLAMMATORY MARKERS AS PREDICTORS
  7. CLINICAL PREDICTORS
  8. COMPOSITE PREDICTIVE INDICES
  9. ENDOSCOPIC INDICES
  10. PATHOLOGICAL FEATURES PREDICTIVE OF RELAPSE
  11. PREDICTION OF POST-SURGICAL RECURRENCE
  12. DISCUSSION
  13. References

Currently, the therapeutic end-point in the treatment of Crohn's disease is the remission of symptoms, but recent data confirm that mucosal inflammation may continue in the absence of symptoms. Furthermore, emerging evidence indicates that such subtle, sub-clinical mucosal inflammation leads to clinical relapse. The assessment of mucosal inflammation has become easier with the availability of faecal calprotectin assay. Current anti-inflammatory therapy often leaves low-grade mucosal inflammation untreated, and therefore recurrent relapses occur. We need to investigate whether the therapeutic end-point of anti-inflammatory medications needs to be more rigorous and to aim at complete mucosal healing, confirmed by the normalization of mucosal inflammatory markers such as faecal calprotectin concentrations. Immunosuppressive therapy with azathioprine/ 6-mercaptopurine currently offers the best mucosal healing treatment with reduction of relapses, but newer biological agents might offer less toxic therapy. Clinical trials to test the feasibility and efficacy of such a paradigm shift in the medical management of Crohn's disease are now warranted.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MUCOSAL MARKERS OF INFLAMMATION PREDICTIVE OF RELAPSE
  5. INTESTINAL PERMEABILITY AS PREDICTOR
  6. SYSTEMIC INFLAMMATORY MARKERS AS PREDICTORS
  7. CLINICAL PREDICTORS
  8. COMPOSITE PREDICTIVE INDICES
  9. ENDOSCOPIC INDICES
  10. PATHOLOGICAL FEATURES PREDICTIVE OF RELAPSE
  11. PREDICTION OF POST-SURGICAL RECURRENCE
  12. DISCUSSION
  13. References

Crohn's disease affects 50 per 100 000 population in the UK, and there has been a well-documented five-fold rise in incidence in Northern Europe since the 1950s.1–3 The disease is characterized by relapsing and remitting chronic intestinal inflammation, and it is estimated that 50% of patients will require surgery within 5 years of diagnosis.4 It is estimated that, after 5 years, 15–20% of patients will be disabled.5 Although mortality ascribed to Crohn's disease is now low, morbidity is considerable, particularly in the young, in whom growth, education, employment potential and social well-being all suffer. Thirty to sixty per cent of patients with Crohn's disease who attain medically induced remission will relapse within 1 year.6, 7 The principal objective of therapy is the reduction of morbidity, perhaps by attenuation of tissue destruction, and improvement in the quality of life.

The target of most medical therapy is the improvement of clinical disease activity, most widely estimated using the Crohn's Disease Activity Index (CDAI), developed during the 1970s by a panel of gastroenterologists in North America.8 A pre-defined CDAI value may be used as an inclusion criterion for recruitment into a trial of medical therapy, and the efficacy of a candidate drug may be measured by either the magnitude of fall in CDAI (often quite a modest drop by 70 points), or the number of patients achieving remission, as defined by CDAI < 150.9 However, CDAI has its limitations as it is subjective and correlates poorly with endoscopic severity.10 Our data, using whole-gut lavage fluid (WGLF) analysis, show the limitations of CDAI in patients with fibrous strictures, in psychologically disturbed patients, in patients with ileostomy and in children.11 Similarly, faecal excretion of 111In-labelled granulocytes showed the presence of `smouldering' mucosal inflammation when the patient was considered to be in clinical remission in a considerable proportion of patients.12 If patients in clinical `remission' continue to demonstrate low-grade inflammatory activity in the mucosa, the concept of clinical remission as a therapeutic end-point may be challenged. In the way of an analogy, we know that clinical end-points are unsatisfactory in bronchial asthma and clinical relief alone will leave considerable disease activity untreated. The peak expiratory flow rate (i.e. a direct bronchial assessment) is accepted as a better therapeutic end-point. However, it is more difficult to obtain a direct measure of intestinal inflammatory activity to use in routine clinical practice.

The ability to predict relapse based on continuing mucosal inflammation might allow the optimization of treatment in high-risk patients, and thus the attenuation of tissue damage and reduction of associated comorbidity. In addition, the power of clinical trials aimed at the maintenance of remission could be improved considerably by the identification of an appropriate patient group at high risk of relapse. In this article, we review evidence regarding the prediction of relapse in Crohn's disease using traditional systemic markers of inflammation as well as new gut-derived inflammatory markers. Reliable identification of such markers of relapse might alter our current goalposts of therapy, and considerable progress in this area has recently been made.

MUCOSAL MARKERS OF INFLAMMATION PREDICTIVE OF RELAPSE

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MUCOSAL MARKERS OF INFLAMMATION PREDICTIVE OF RELAPSE
  5. INTESTINAL PERMEABILITY AS PREDICTOR
  6. SYSTEMIC INFLAMMATORY MARKERS AS PREDICTORS
  7. CLINICAL PREDICTORS
  8. COMPOSITE PREDICTIVE INDICES
  9. ENDOSCOPIC INDICES
  10. PATHOLOGICAL FEATURES PREDICTIVE OF RELAPSE
  11. PREDICTION OF POST-SURGICAL RECURRENCE
  12. DISCUSSION
  13. References

More than a decade ago, Saverymuttu et al. described the use of 111In-labelled granulocyte scanning to quantify intestinal mucosal inflammation, both by dynamic imaging and by faecal excretion.13, 14 A large proportion of Crohn's disease patients in clinical remission showed evidence of continuing mucosal inflammation.12 On the basis of this observation, Saverymuttu et al. raised the question of whether treating sub-clinical inflammation would alter the natural history of the disease. Studies from a number of different research teams have now provided persuasive evidence that such `smouldering' sub-clinical mucosal inflammation after medically induced remission is associated with subsequent relapse within a year. A number of different techniques have recently been used to document such low-grade mucosal inflammation. The principle is, however, the same as radiolabelled white cell excretion — for example, it has been shown that 4-day faecal calprotectin and daily excretions correlate significantly with the 4-day faecal excretion of 111In-labelled white cells, and this significant correlation is maintained when single stool concentrations of faecal calprotectin are used.15

Intestinal mucosal cytokines

Schreiber et al. investigated the expression of the pro-inflammatory cytokines, interleukin-1β (IL1β) and tumour necrosis factor-α (TNFα), in cultured lamina propria mononuclear cells in 137 patients after steroid-induced remission (CDAI < 150).16 After 1% pokeweed mitogen stimulation, TNFα concentrations were successfully measured in 125 and IL1β concentrations were successfully measured in 119 out of 137 patients. Patients whose stimulated lamina propria cells secreted high concentrations of TNFα or IL1β had a high risk of the development of an inflammatory relapse during the following year, but the associated clinical and serological parameters were not of predictive value over a year. Patients in whom the secretion of TNFα was > 70 pg/mL had a significantly higher risk of relapse within a year compared with those with a lower secretion of TNFα [hazard ratio, 0.43; 95% confidence interval (CI), 0.24–0.76; P=0.004]. The median time to relapse was 79 days (interquartile range, 44–132 days) among those with high TNFα secretion (> 70 pg/mL) and 200 days (interquartile range, 90–287 days) among those with low TNFα secretion. The risk of relapse increased continuously over a wide range of TNFα concentrations, and patients in whom the secretion of TNFα by lamina propria mononuclear cells reached 150 pg/mL or more had only a 30% chance of staying in remission. The concentration of IL1β secreted by lamina propria cells had a similar association with the risk of relapse. The median time to relapse was 84 days (interquartile range, 34–132 days) amongst patients with high concentrations of secreted IL1β (≥ 75 pg/mL), compared with 216 days (interquartile range, 90–289 days) for those with lower concentrations of IL1β (P=0.008). Seventy-three per cent of patients with IL1β < 75 pg/mL stayed in remission for 1 year. This research provides important data on relapse in a homogeneous group of patients after steroid therapy and may indicate those suitable for further immunosuppressive therapy. This study clearly shows that persistent subtle mucosal inflammation after steroid-induced clinical remission leads to subsequent relapse. It is possible that steroid-induced remission might be immunologically `incomplete', despite providing relief from clinical symptoms in a significant proportion of patients.

Faecal calprotectin

Calprotectin is a calcium-binding protein comprising up to 60% of the total cytosolic protein content of neutrophils, and is stable during intestinal transit. It can be assayed easily in faeces and it is useful as a quantitative objective measure of intestinal inflammation. Tibble et al. assessed the value of the faecal calprotectin assay for the prediction of relapse in a group of 43 inactive Crohn's disease patients. 17 Patients with chronic active Crohn's disease requiring > 5 mg prednisolone per day were excluded. A single stool assay by enzyme-linked immunoabsorbent assay and a cut-off threshold of 50 mg/L using receiver operating characteristics gave the best combination of sensitivity and specificity for all inflammatory bowel disease patients. At this threshold, the sensitivity for the prediction of relapse was 90% with a specificity of 83% (Table 1). The relative risk of relapse in patients with an elevated faecal calprotectin level was 10.6 (95% CI, 2.5–45.8; P=0.002). Fifty-eight per cent of the Crohn's disease patients relapsed over a 12-month follow-up period. Median faecal calprotectin concentrations in the Crohn's disease patients who relapsed was 122 mg/L (95% CI, 98–229 mg/L), significantly higher (P < 0.0001) than those who did not relapse (median, 42 mg/L; 95% CI, 31–49 mg/L). Erythrocyte sedimentation rate and C-reactive protein concentrations did not predict relapse. Faecal calprotectin is a non-invasive and simple method to assess prognosis and may have clinical applications. It is especially attractive for use in paediatric patients. A high faecal calprotectin in clinically inactive Crohn's disease patients might indicate a stage of active `smouldering' mucosal inflammation, which progresses to cause eventual clinical relapse of the disease. True confirmation of this hypothesis will require sequential measurement of faecal calprotectin in patients with clinically quiescent Crohn's disease.

Table 1.  . Sensitivity and specificity of direct and indirect markers of mucosal inflammation in the prediction of clinical relapse of quiescent Crohn's disease Thumbnail image of

Gut lavage inflammatory markers

Whole-gut lavage fluid analysis is an assessment of mucosal immunity and inflammation. The WGLF immunoglobulin G correlates well with CDAI and is thus a marker of disease activity. Our group has identified WGLF IL1β and IL8 as factors predictive of relapse in a group of 54 patients with inactive Crohn's disease.18 Inactive disease was rigorously defined as a CDAI of less than 150 and a WGLF immunoglobulin G < 10 μg/mL. A total of 54 inactive Crohn's disease patients were followed up for 1 year. Patients with a WGLF IL1β concentration of > 12 pg/mL had a significantly greater chance of relapsing than those with a WGLF IL1β concentration up to 12 pg/mL (P < 0.004). The sensitivity of elevated WGLF IL1β in predicting relapse was 92% with a specificity of 54%. The relative risk of relapse in Crohn's disease patients with an elevated IL1β was 11.0. Similarly, patients with a WGLF IL8 concentration higher than the reference range, i.e. > 60 pg/mL, had a significantly greater risk of clinical relapse compared with those with a WGLF IL8 within the reference range (P < 0.02; relative risk, 4.2). Stepwise multiple regression analysis identified only the WGLF IL1β concentration as a significant independent variable. Ghosh et al., in a similar study, reported as an abstract, also identified granulocyte elastase (a marker of the presence of luminal neutrophils) in WGLF as predictive of relapse.21 It is likely that the detection of cytokines or granulocyte-derived enzymes indicates sub-clinical mucosal inflammation, despite the resolution of clinical symptoms and mucosal protein loss.

INTESTINAL PERMEABILITY AS PREDICTOR

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MUCOSAL MARKERS OF INFLAMMATION PREDICTIVE OF RELAPSE
  5. INTESTINAL PERMEABILITY AS PREDICTOR
  6. SYSTEMIC INFLAMMATORY MARKERS AS PREDICTORS
  7. CLINICAL PREDICTORS
  8. COMPOSITE PREDICTIVE INDICES
  9. ENDOSCOPIC INDICES
  10. PATHOLOGICAL FEATURES PREDICTIVE OF RELAPSE
  11. PREDICTION OF POST-SURGICAL RECURRENCE
  12. DISCUSSION
  13. References

An increase in intestinal permeability, which may be a surrogate marker of intestinal inflammation, was one of the first suggested predictors of clinical relapse in Crohn's disease. Wyatt et al. found the lactulose/mannitol test of intestinal permeability to be predictive of relapse in 72 patients with quiescent Crohn's disease, as defined by CDAI, with a sensitivity of 81% at 1 year.20 The association of intestinal permeability with relapse has been replicated in three studies. D'Inca et al. identified the lactulose/mannitol ratio as an independent variable in a group of 130 patients. In this group of patients with serial intestinal permeability, blood and clinical assessments, intestinal permeability had a sensitivity of 53.3% and a specificity of 84.6% for the prediction of relapse in the forthcoming 4 months.19 Tibble et al. found that the lactulose/rhamnose assessment of intestinal permeability had a sensitivity of 84% and a specificity of 64% for the identification of relapse in a group of 43 quiescent Crohn's disease patients.17 Our group has identified the lactulose/rhamnose ratio as an independent variable able to predict relapse over a 1-year follow-up period in a group of 50 inactive Crohn's disease patients with a sensitivity of 89% and specificity of 76%.22 A smaller study using 51Cr ethylenediaminetetra-acetate has not reproduced the association. Intestinal permeability does not predict endoscopic relapse following ileocolonic resection,23 although there has been one reported case of abnormal intestinal permeability preceding the onset of Crohn's disease in a subject with a family history.24 Concomitant medication, especially non-steroidal anti-inflammatory drugs, alcohol and smoking, are all relevant in the assessment of intestinal permeability. It is likely that intestinal permeability will only be of use in predicting relapse in patients with small bowel Crohn's disease, as less than 50% of patients with active colonic Crohn's disease have an increased intestinal permeability.

SYSTEMIC INFLAMMATORY MARKERS AS PREDICTORS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MUCOSAL MARKERS OF INFLAMMATION PREDICTIVE OF RELAPSE
  5. INTESTINAL PERMEABILITY AS PREDICTOR
  6. SYSTEMIC INFLAMMATORY MARKERS AS PREDICTORS
  7. CLINICAL PREDICTORS
  8. COMPOSITE PREDICTIVE INDICES
  9. ENDOSCOPIC INDICES
  10. PATHOLOGICAL FEATURES PREDICTIVE OF RELAPSE
  11. PREDICTION OF POST-SURGICAL RECURRENCE
  12. DISCUSSION
  13. References

C-reactive protein

C-reactive protein, so called because it reacts with the C-polysaccharide of pneumococci, is one of the acute phase proteins. It is part of the α-globulin fraction and is synthesized in the liver in response to a number of mediators, including IL6 and IL1β. Andre et al. described five patients in whom C-reactive protein rose prior to clinical relapse.25 Boirivant et al. longitudinally assessed C-reactive protein in 101 patients, of whom 48 were initially in remission.26 There were no differences in the relapse rates in the first year, but patients who had been in remission for the previous 12 months and had a persistently elevated C-reactive protein had a significantly higher relapse rate in the second year than those who did not. One-third of patients with active disease and one-third of patients in remission had a raised C-reactive protein. In children, C-reactive protein has been proposed to be of prognostic value, with a greater sensitivity and specificity than in adult Crohn's disease patients.27 Although elevated C-reactive protein might predict relapse, most patients in remission who subsequently relapse have normal C-reactive protein concentrations.

Interleukin-6

Louis et al. described an elevated serum IL6 as an independent variable predictive of relapse in a study of 36 Crohn's disease patients in remission.28 The sensitivity was 73% with a specificity of 96%. A further smaller study of 19 patients with steroid-induced remission quoted a 100% sensitivity and specificity of serum IL6 for the prediction of relapse at 5 months.29 However, in a retrospective analysis of 36 patients, only soluble IL2 receptor, but not serum IL6, was predictive of relapse. Elevated serum IL6 was associated with a more frequently relapsing pattern of disease, but was not a predictive factor.30

Soluble interleukin-2 receptor

Louis et al. assessed the concentration of soluble IL2 receptor in 58 patients with Crohn's disease.31 Of these, 29 patients were in remission, as defined by CDAI, and were followed longitudinally for 1 year. Soluble IL2 receptor, C-reactive protein, erythrocyte sedimentation rate, relative lymphocytosis, fibrinogen and α-1 glycoprotein were assessed, together with a number of historical and clinical parameters. Of the 29 patients in remission, nine relapsed during follow-up and the soluble IL2 receptor was shown to be elevated in eight of the nine cases, although an additional seven patients who had an elevated level at entry did not relapse during the follow-up period. This gave soluble IL2 receptor a sensitivity of 88% and a specificity of 65% for the prediction of relapse within 1 year.

Other serum markers

A large number of serum markers of relapse have been proposed and investigated. These include α-1 glycoprotein, α-1 antitrypsin, γ-globulin, orosomucoid, fibrinogen and relative lymphocytosis, but the sensitivity and specificity are relatively low and the data are inconsistent.

Overall, serum markers are convenient to measure, but have a relatively low sensitivity and specificity for the prediction of relapse. Few patients in clinical remission have elevated serum inflammatory markers. There is a disparity between systemic and mucosal inflammation, with serum markers being a surrogate marker of mucosal events. In patients in clinical remission, the frequency of abnormal serum inflammatory parameters is very low and therefore not clinically useful.32

CLINICAL PREDICTORS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MUCOSAL MARKERS OF INFLAMMATION PREDICTIVE OF RELAPSE
  5. INTESTINAL PERMEABILITY AS PREDICTOR
  6. SYSTEMIC INFLAMMATORY MARKERS AS PREDICTORS
  7. CLINICAL PREDICTORS
  8. COMPOSITE PREDICTIVE INDICES
  9. ENDOSCOPIC INDICES
  10. PATHOLOGICAL FEATURES PREDICTIVE OF RELAPSE
  11. PREDICTION OF POST-SURGICAL RECURRENCE
  12. DISCUSSION
  13. References

The definition of relapse and of disease activity remain central to the issue of the prediction of relapse in Crohn's disease.33 Three major studies have specifically examined clinical markers of relapse in Crohn's disease. The Groupe d'Etudes Therapeutiques des Affections Inflammatoires Digestives (GETAID) group assessed patients in three clinical trials and identified an age less than 25 years, interval since first symptoms of > 5 years, interval since last relapse of less than 6 months and colonic involvement as poor prognostic markers.34 One thousand and eighty-four patients who were entered into the National Co-operative Crohn's Disease Study were analysed by Mekhjian et al.35 In 77 patients who were randomized to the placebo arm, previous surgical removal of all observable disease, absence of perianal disease and a CDAI of less than 200 were favourable prognostic factors, while steroids prior to randomization was a poor prognostic factor. The proportion of patients with colonic disease alone in the National Co-operative Crohn's Disease Study was only 11%, whereas it was greater than 30% in two large European studies.36, 37 Steinhardt et al. investigated 110 patients who were randomized to the placebo group in the European Co-operative Crohn's Disease Study.36 Of these, 53 had quiescent disease and 57 had active disease as defined by CDAI. Favourable predictive factors were a long duration of disease between diagnosis and randomization, normal serum albumin and ileocolonic disease. In contrast, factors that were predictive of a poorer outcome were extensive small bowel and isolated colonic disease, treatment with steroids and bowel resection prior to entry into the study.

Most of the clinical markers of relapse are immutable and, furthermore, a number of studies suffer from a relatively small sample size, heterogeneous populations, differing definitions of relapse and remission and the use of these markers as primary or secondary end-points. In addition, these markers are often not independent variables when assessed with multivariate analysis. Young age has been identified as a marker of relapse,18, 34, 38 but a number of smaller studies have failed to confirm this association.17, 20, 28, 31, 39 The time since last relapse has similarly been identified,18, 34, 39 but the data are conflicting.28, 31 A long interval since diagnosis was identified in a large study,34 but not in smaller cohorts.18, 31, 39 The disease distribution that is most consistently associated with frequent relapse is colonic disease,34, 36, 40 but ileal disease22 and ileocolonic disease40, 41 have also been identified. Perianal disease appears to be an independent marker of a poor prognosis.28, 35, 40, 42 The need for corticosteroids,31, 35, 36, 38 previous surgery36 and the type of presentation41 (inflammatory or obstructive) have also been identified as prognostic markers, but the data are inconsistent. Most of the clinical predictors of relapse are therefore either immutable or inconsistent.

The exception is smoking, the single most important environmental influence conferring a poor prognosis in Crohn's disease. Smoking is not only associated with an increased susceptibility for the disease, but current smoking leads to an increased frequency of relapse, more surgery, more rapid recurrence post-surgery,43, 44 more severe recurrent anastomotic lesions,45 a poorer quality of life46 and an overall greater mortality.47 In addition, there are now prospective data to suggest that the cessation of smoking has a positive impact on the disease course.48 The use of oral contraceptives has been proposed as a poor prognostic marker, although the data are far less convincing.49

Clinical markers of relapse for patients on long-term azathioprine or 6-mercaptopurine have been assessed.50 Whilst on therapy, female gender, an age less than 26 years and a delay in achieving remission of greater than 6 months were poor prognostic markers. When therapy was withdrawn, male gender, an age less than 31 years and a duration of remission of less than 4 years were indicative of a greater chance of relapse.50

COMPOSITE PREDICTIVE INDICES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MUCOSAL MARKERS OF INFLAMMATION PREDICTIVE OF RELAPSE
  5. INTESTINAL PERMEABILITY AS PREDICTOR
  6. SYSTEMIC INFLAMMATORY MARKERS AS PREDICTORS
  7. CLINICAL PREDICTORS
  8. COMPOSITE PREDICTIVE INDICES
  9. ENDOSCOPIC INDICES
  10. PATHOLOGICAL FEATURES PREDICTIVE OF RELAPSE
  11. PREDICTION OF POST-SURGICAL RECURRENCE
  12. DISCUSSION
  13. References

A number of both clinical and laboratory indices have been constructed to identify patients at high risk of relapse. An example is the Italian (Brignola) Index.39 This was developed by discriminant analysis and is calculated thus: – 3.5 + (erythrocyte sedimentation rate × 0.03) + (acid α-1-glycoprotein × 0.013) + (α-2-glycoprotein). Although the resulting number was found to be predictive of relapse, it is complex and unlikely to be used outside clinical trials. Similarly developed indices include the Crohn's Activity Group Scale,38 the GETAID Prognostic Score34 and the Belgian (Louis) Prognostic Models.31 All require complicated calculations and often add little to the sensitivity and specificity of the single markers.

ENDOSCOPIC INDICES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MUCOSAL MARKERS OF INFLAMMATION PREDICTIVE OF RELAPSE
  5. INTESTINAL PERMEABILITY AS PREDICTOR
  6. SYSTEMIC INFLAMMATORY MARKERS AS PREDICTORS
  7. CLINICAL PREDICTORS
  8. COMPOSITE PREDICTIVE INDICES
  9. ENDOSCOPIC INDICES
  10. PATHOLOGICAL FEATURES PREDICTIVE OF RELAPSE
  11. PREDICTION OF POST-SURGICAL RECURRENCE
  12. DISCUSSION
  13. References

Endoscopic recurrence is seen in the majority of patients following ileocolonic resection for Crohn's disease.51 This is generally pre-anastomotic and there is evidence that the severity of macroscopic changes relates to subsequent clinical relapse. Rutgeerts et al. analysed data from 89 patients who underwent resection for ileocolonic Crohn's disease, and identified the severity of early post-operative lesions on ileoscopy as the best predictive factor for clinical relapse.52 These findings are an extrapolation of earlier similar work and provide similar results.53 Patients with severe changes in the neo-terminal ileum, such as deep ulceration and mucosal involvement, were found to relapse more frequently than those with mild changes, such as mucosal erythema. Aphthous ulceration in the neo-terminal ileum was independent of clinical relapse. Similar findings were observed in a smaller Scandinavian study.54 deJong et al. did not observe an association between post-operative endoscopic lesions and subsequent clinical relapse, but the endoscopic definitions were more poorly defined and the follow-up was shorter.55 Lorenz-Myer et al. colonoscoped 130 patients in the European Co-operative Crohn's Disease Study and found that colonoscopic appearances per se were not of prognostic significance.56 Endoscopic appearances do not correlate with symptoms or clinical disease activity.57 However, unlike adult gastroenterologists, paediatric gastroenterologists have often sought mucosal healing rather than clinical remission. In an interesting study, TNFα-secreting cells were found to be increased in the mucosa of inflamed intestine, with especially high levels observed in Crohn's disease.58 No relationship existed between histological healing and the frequency of TNFα-secreting cells. The frequency of TNFα-secreting cells, was, however, not reduced after steroid or enteral nutrition. Unlike the post-surgical situation, endoscopic or histological definition of mucosal healing using current scoring systems is likely to be of limited prognostic value after medical induction of remission.

PATHOLOGICAL FEATURES PREDICTIVE OF RELAPSE

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MUCOSAL MARKERS OF INFLAMMATION PREDICTIVE OF RELAPSE
  5. INTESTINAL PERMEABILITY AS PREDICTOR
  6. SYSTEMIC INFLAMMATORY MARKERS AS PREDICTORS
  7. CLINICAL PREDICTORS
  8. COMPOSITE PREDICTIVE INDICES
  9. ENDOSCOPIC INDICES
  10. PATHOLOGICAL FEATURES PREDICTIVE OF RELAPSE
  11. PREDICTION OF POST-SURGICAL RECURRENCE
  12. DISCUSSION
  13. References

Nagel et al. examined the resection margins of 29 patients with a scanning electron microscope and reported a triad of abnormalities present in macroscopic and histological normal sections.59 They postulated that endoscopic recurrence was proportional to the number of early lesions seen at surgery and had little bearing on prognosis. Heinmann et al. found that inflammation at the resection margins was not of predictive value in those with ileostomy or a single resection, although it was in those with multiple anastomoses.60 In addition, Heinmann et al. also suggested that a low pre-operative lymphocyte count was predictive of relapse in 65 patients prospectively assessed prior to surgery.61 Markowitz et al. identified epithelioid granulomata in recto-sigmoid biopsies as being predictive of more extensive Crohn's disease and an increased need for surgery, although clinical disease activity scores and hospitalization were identical.62

PREDICTION OF POST-SURGICAL RECURRENCE

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MUCOSAL MARKERS OF INFLAMMATION PREDICTIVE OF RELAPSE
  5. INTESTINAL PERMEABILITY AS PREDICTOR
  6. SYSTEMIC INFLAMMATORY MARKERS AS PREDICTORS
  7. CLINICAL PREDICTORS
  8. COMPOSITE PREDICTIVE INDICES
  9. ENDOSCOPIC INDICES
  10. PATHOLOGICAL FEATURES PREDICTIVE OF RELAPSE
  11. PREDICTION OF POST-SURGICAL RECURRENCE
  12. DISCUSSION
  13. References

Relapse rates

Extensive reviews have been published on the subject of post-surgical recurrence of Crohn's disease,63, 64 as the identification of the risk factors for relapse and the modification of the natural history by medical therapy are important aims following surgical resection.63, 64 The frequency of post-surgical relapse has been estimated to be 37% of patients at 3 years.65 Lock et al. estimated the risk of recurrence requiring surgery to be 3.9% per annum with a cumulative risk of 41.8% at 15 years.66 Greenstein et al. estimated the 15-year re-operation rate to be 89% and the overall clinical recurrence to be 94% in a group of patients with colitis and ileocolitis.67

Clinical indices

Smoking adversely influences the recurrence of Crohn's disease following removal of observable disease. Smokers require repeated surgery sooner43 and have more severe recurrent anastomotic lesions.47 A number of other clinical features have been assessed, but the data are more inconsistent. Whelan et al. assessed 615 patients with Crohn's disease, 438 of whom had undergone resective surgery.68 They identified that ileocolonic involvement needed a second resection most frequently. The second resection rates in those with only ileal and colonic distributions were the same. Intestinal fistula and perianal disease as indications for surgery were associated with the poorest outcome.

Binder et al. identified the presence of intestinal fistula and the need for steroid therapy as independent risk factors for a poorer prognosis in a stringently selected group of patients with isolated ileal Crohn's disease.69 Holzheimer et al. examined whether six clinical variables influenced clinical recurrence after first resection for Crohn's disease.70 Perforation as an indication for surgery and the presence of post-operative complications were identified as poor prognostic indicators. Rutgeerts et al. identified pre-operative disease activity, the indication for surgery and the number of resections to be of predictive value.52 Griffiths et al. identified ileocolonic disease, a failure of medical therapy as an indication for surgery and a long duration of pre-operative disease as predictors of early post-operative recurrence in children,71 and similar findings were published by Basilisco et al.72 Heinmann et al. identified the need for multiple surgical resections as a poor predictive factor and felt that ileostomy was associated with a significantly lower recurrence rate than the use of single or multiple anastomoses.60 Post et al. identified juvenile onset, proximal small bowel disease and a fistulating disease type as risk factors for a second operation.73 Besnard et al. identified the extent of involvement prior to surgery as the only decisive factor in predicting recurrence in 30 children undergoing surgery for Crohn's disease.74 Pallone et al. found that the phenotype of the pre-operative disease predicted the post-operative course.75

The major clinical factor influencing post-operative recurrence is smoking. Pre-operative disease activity, fistulating or perianal disease as an indication for surgery and the need for pre-operative steroids may also be weaker predictive markers.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MUCOSAL MARKERS OF INFLAMMATION PREDICTIVE OF RELAPSE
  5. INTESTINAL PERMEABILITY AS PREDICTOR
  6. SYSTEMIC INFLAMMATORY MARKERS AS PREDICTORS
  7. CLINICAL PREDICTORS
  8. COMPOSITE PREDICTIVE INDICES
  9. ENDOSCOPIC INDICES
  10. PATHOLOGICAL FEATURES PREDICTIVE OF RELAPSE
  11. PREDICTION OF POST-SURGICAL RECURRENCE
  12. DISCUSSION
  13. References

There are a number of reproducible clinical markers that identify the group of Crohn's disease patients in remission who are at high risk of forthcoming relapse. These include young age at diagnosis and colonic disease distribution. These immutable factors are good markers within cohorts, but poorly relate to individuals and are not amenable to therapeutic intervention. In addition, there have also been a number of serum markers proposed as prognostic indicators. Although applicable to individuals, these, especially C-reactive protein, lack the sensitivity and specificity that would enable effective patient management and, in reality, C-reactive protein is a marker of active disease that is not often elevated in patients in remission. Serum soluble IL2 receptor and IL6 are superior to C-reactive protein, but remain surrogate markers of intestinal inflammation. Of the systemic markers of inflammation predictive of relapse in Crohn's disease, soluble IL2 receptor and IL6 are probably the most promising. Gut-derived markers are a more direct assessment of intestinal inflammation and the data available thus far indicate that these might have the sensitivity and specificity to enable a different approach to be used for the treatment of quiescent inflammatory bowel disease. Faecal calprotectin is likely to be the most convenient of these gut-derived markers, but further confirmation of the initial observations is required in a larger number of patients stratified according to smoking status and different treatment groups. Most of the current data have been derived from steroid-induced remission. Patients with azathioprine-induced remission may have abnormal intestinal permeability without early relapse.22 Whether treatment of this subtle mucosal inflammation (and not just clinically ill patients) will reduce relapse requires further study and may represent a paradigm shift in the approach to the medical treatment of Crohn's disease.

Mucosal remission as the therapeutic end-point

To date, therapeutic end-points have been quantified by symptom scores, either from a clinical history in the out-patient clinic or more formally by disease activity indices within clinical trials. However, the assessment of clinical disease activity has limitations. It does not provide prognostic information, is relatively less useful when the patient is in remission and symptoms also correlate poorly with endoscopic and histological measures of disease activity. The target of anti-inflammatory therapy in Crohn's disease is attenuation of intestinal inflammation. There are a number of situations in which symptom scores may be far in excess of mucosal inflammatory activity, for example with diarrhoea following ileal resection. Likewise, ongoing but low-grade and subtle inflammation in patients in remission, although of prognostic importance, cannot be assessed by symptoms. The proposition to use mucosal remission as a therapeutic end-point has a number of theoretical advantages. It is able to identify a population of patients who, despite being in clinical remission, are at a high risk of forthcoming relapse. It removes some difficulties inherent with disease heterogeneity, psychosocial interaction and post-operative assessment. In terms of a prognostic marker, assessment of mucosal inflammatory activity is more applicable to an individual and more amenable to therapeutic intervention than, for example, the immutable clinical markers of relapse, such as age or disease distribution.

Systemic markers of inflammation as predictors of relapse

The existing systemic markers of relapse are generally inferior to gut-derived markers, although soluble IL2 receptor and serum IL6 are superior to C-reactive protein. Comparative data from patients in remission undergoing simultaneous gut lavage and serum analysis show very few serum abnormalities when gut lavage indicates inactive disease. Only 10% of 30 inactive Crohn's disease patients had abnormal serum inflammatory markers, as opposed to 58% who had abnormal WGLF inflammatory markers.32 Therefore, serum markers will at best identify a small sub-group of quiescent Crohn's disease patients at risk of relapse and will be of limited use.

Therapeutic strategy based on mucosal remission

The ability to predict relapse with faecal calprotectin gives a non-invasive method of identifying a group of patients who are at high risk of forthcoming relapse. The treatment of such patients to prevent relapse might attenuate morbidity and improve the quality of life. Azathioprine might be a potential therapeutic choice in this situation, as it is a widely used and efficacious agent for the maintenance of remission in Crohn's disease. Ten to fifteen per cent of patients are unable to continue therapy with azathioprine due to adverse events, but the incidence of adverse events might be even higher.76 Although many of these side-effects, such as nausea, are not life threatening, serious adverse events, such as myelosuppression and liver problems, raise concern about treatment in those who may not require such medication. Therefore, additional treatments in asymptomatic patients must attempt to avoid the consequences of long- and short-term immunosuppression. Strategies, such as the analysis of thiopurine methyltransferase activity, may be employed to minimize adverse events and must be an integral part of such treatments. Further knowledge of pharmacogenetic mechanisms will help to minimize adverse events in the future. In addition, there is anecdotal evidence that lower doses of maintenance therapy may be efficacious in this situation in some patients. We propose, as an experimental assessment of this strategy, that patients brought into remission with a course of reducing steroids should have sub-clinical mucosal inflammation assessed by faecal calprotectin, and those with increased concentrations should be randomized to receive either a maintenance agent or no treatment. Careful assessment of drug side-effects and relapse rates is mandatory. Analysis of low-dose azathioprine regimens may also be appropriate in this setting. However, less toxic novel agents may be better suited to the role than the existing available medications.

Future work — how best to define mucosal remission?

Methods of assessing mucosal remission include endoscopy with or without conventional histology, the analysis of gut-derived cytokines either via biopsy or whole-gut lavage, indirectly by the assessment of intestinal permeability or by the assessment of faecal proteins, such as calprotectin. Endoscopic appearance or histological scores correlate poorly with clinical disease activity. The analysis of faecal calprotectin has some advantages. Samples are easily obtained and the method is not invasive and does not involve ionizing radiation. The analysis is simple and inexpensive. It is possible that more sophisticated methods of detection of mucosal inflammation may become available in the future. The assessment of this strategy requires a thorough clinical evaluation with well-conducted and rigorously controlled clinical trials. A large number of patients, stratified according to smoking status and different treatment sub-groups (such as steroids or azathioprine), require further study to confirm the observations made predominantly on patients attaining corticosteroid-induced remission. Only then may it be possible to change our approach to patients with inactive Crohn's disease. The newly identified genotype NOD2/CARD15, associated with Crohn's disease, and other genotypes which will inevitably be identified, might provide other means of prognostication. It would, however, appear inevitable that the resolution of clinical symptoms alone will cease to be the only therapeutic end-point in Crohn's disease. The current practice of patients having to earn immunosuppressive therapy by demonstrating multiple clinical relapses is inefficient and may promote excessive tissue destruction.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MUCOSAL MARKERS OF INFLAMMATION PREDICTIVE OF RELAPSE
  5. INTESTINAL PERMEABILITY AS PREDICTOR
  6. SYSTEMIC INFLAMMATORY MARKERS AS PREDICTORS
  7. CLINICAL PREDICTORS
  8. COMPOSITE PREDICTIVE INDICES
  9. ENDOSCOPIC INDICES
  10. PATHOLOGICAL FEATURES PREDICTIVE OF RELAPSE
  11. PREDICTION OF POST-SURGICAL RECURRENCE
  12. DISCUSSION
  13. References
  • 1
    Barton JR, Gillon S, Ferguson A. Incidence of inflammatory bowel disease in Scottish children between 1968 and 1983; marginal fall in ulcerative colitis, three fold rise in Crohn's disease. Gut 1989; 30: 61822.
  • 2
    Rose JDR, Roberts GM, Williams G, Mayberry JF, Rhodes J. Cardiff Crohn's disease jubilee: the incidence of over 50 years. Gut 1988; 29: 34651.
  • 3
    Armitage E, Drummond H, Ghosh S, Ferguson A. Incidence of juvenile-onset Crohn's disease in Scotland [letter]. Lancet 1999; 353: 14967.
  • 4
    Rutgeerts PJ. Prevention of early recurrence of Crohn's disease after ileal resection with ileocolonic anastomosis. Eur J Gastroenterol Hepatol 1994; 6: 1136.
  • 5
    Binder V, Hendriken C, Kreiner S. Prognosis in Crohn's disease — based on results from a regional patient group from the county of Copenhagen. Gut 1985; 26: 14650.
  • 6
    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: 699706.
  • 7
    Greenberg GR, Feagan BG, Martin F, et al.Oral budesonide as maintenance treatment for Crohn's disease: a placebo controlled dose-ranging study. Gastroenterology 1996; 110: 4551.
  • 8
    Best WR, Becktel JM, Singleton JW, Kern F. Development of a Crohn's disease index; National Cooperative Crohn's Disease Study. Gastroenterology 1976; 70: 43944.
  • 9
    Hodgson HJF, Bhatti M. Assessment of disease activity in ulcerative colitis and Crohn's disease. Inflammatory Bowel Dis 1995; 1: 11734.
  • 10
    Cellier C, Sahmound T, Froguel E, et al.Correlations between clinical activity, endoscopic severity and biological parameters in colonic or ileocolonic Crohn's disease — a prospective multicentre study of 121 cases. Gut 1994; 35: 2315.
  • 11
    Acciuffi S, Ghosh S, Ferguson A. Strengths and limitations of the Crohn's disease activity index, revealed by an objective gut lavage test of gastrointestinal protein loss. Aliment Pharmacol Ther 1997; 10: 3216.
  • 12
    Saverymuttu SH. Clinical remission in Crohn's disease — assessment using faecal 111-In granulocyte excretion. Digestion 1986; 33: 749.
  • 13
    Saverymuttu SH, Camilleri M, Rees H, Lavender JP, Hodgson HJF, Chadwick VS. Indium 111-granulocyte scanning in the assessment of disease extent and disease activity in inflammatory bowel disease. Gastroenterology 1986; 90: 11218.
  • 14
    Saverymuttu SH, Peters AM, Lavender JP, Pepys MB, Hodgson HJF, Chadwick VS. Quantitative faecal Indium 111-labeled leukocyte excretion in the assessment of disease in Crohn's disease. Gastroenterology 1983; 85: 13339.
  • 15
    Roseth AG, Schmidt PN, Fagerhol MK. Correlation between faecal excretion of indium-111-labelled granulocytes and calprotectin, a granulocyte marker protein, in patients with inflammatory bowel disease. Scand J Gastroenterol 1999; 34: 504.
  • 16
    Schreiber S, Nikolaus S, Hampe J, et al.Tumour necrosis factor α and interleukin 1β in relapse of Crohn's disease. Lancet 1999; 353: 45961.
  • 17
    Tibble J, Sigthorsson G, Bridger S, Fagerhol MK, Bjarnason I. Surrogate markers of intestinal inflammation are predictive of relapse in patients with inflammatory bowel disease. Gastroenterology 2000; 119: 1522.
  • 18
    Arnott IDR, Drummond H, Ghosh S. Gut mucosal secretion of interleukin 1β and interleukin-8 predicts relapse in clinically inactive Crohn's disease. Dig Dis Sci 2001; 46: 4029.
  • 19
    D'Inca R, Di Leo V, Corrao G, et al.Intestinal permeability test as a predictor of clinical course in Crohn's disease. Am J Gastroenterol 1999; 94: 295660.
  • 20
    Wyatt J, Vogelsang H, Hubl W, Waldhoer T, Lochs H. Intestinal permeability and the prediction of relapse in Crohn's disease. Lancet 1993; 341: 14379.
  • 21
    Ghosh S, Acciuffi S, Ferguson A. Prognostic value of whole gut lavage fluid analysis in Crohn's disease. Gut 1996; 38(Suppl. 1): F231F231.
  • 22
    Arnott IDR, Kingstone K, Ghosh S. Abnormal intestinal permeability predicts relapse in patients with inactive Crohn's disease. Scand J Gastroenterol 2000; 35: 11639.
  • 23
    Klein O, Houdret N, Desreumaux P, et al.Intestinal permeability is not predictive of endoscopic recurrence in patients operated on for Crohn's disease. Gastroenterology 1997; 112: A1016A1016(Abstract).
  • 24
    Irvine EJ, Marshall JK. Increased intestinal permeability precedes the onset of Crohn's disease in a subject with familial risk. Gastroenterology 2000; 119: 17404.
  • 25
    Andre C, Descos I, Vignal J, Gillon J. C-reactive protein as a predictor of relapse in asymptomatic patients with Crohn's disease. Scott Med J 1983; 28: 269.
  • 26
    Boirivant M, Leoni M, Taricotti D, Fais S, Squarcia O, Pallone F. The clinical significance of serum C reactive protein levels in Crohn's disease. J Clin Gastroenterol 1988; 10: 4015.
  • 27
    Campbell CA, Walker-Smith JA, Hindocha P, Adinolfi M. Acute phase proteins in chronic inflammatory bowel disease in childhood. J Pediatr Gastroenterol Nutr 1982; 1: 193200.
  • 28
    Louis E, Belaiche J, Van Kemseke C, et al.A high serum concentration of interleukin-6 is predictive of relapse in quiescent Crohn's disease. Eur J Gastroenterol Hepatol 1997; 9: 93944.
  • 29
    Reinisch W, Gasche C, Tillinger W, et al.Clinical relevance of serum interleukin-6 in Crohn's disease: Single point measurements, therapy monitoring and prediction of clinical relapse. Am J Gastroenterol 1999; 94: 215664.
    Direct Link:
  • 30
    Van Kemseke C, Belaiche J, Louis E. Frequently relapsing Crohn's disease is characterized by persistent elevation in interleukin-6 and soluble interleukin-2 receptor serum levels during remission. Int J Colorectal Dis 2000; 15: 20610.
  • 31
    Louis E, Belaiche J, Van Kemseke C, Schaaf N, Mahieu P, Mary JY. Soluble interleukin-2 receptor in Crohn's disease, assessment of disease activity and prediction of relapse. Dig Dis Sci 1994; 40: 17506.
  • 32
    Arnott IDR, Drummond HE, Ghosh S. Frequency of continuing mucosal inflammation in clinically inactive Crohn's disease. Scott Med J 2001; 46: 1369.
  • 33
    Sachar DB. Predicting relapses in Crohn's disease. Dig Dis Sci 1987; 32: 2068.
  • 34
    Sahmound T, Hoctin-Boes G, Modigliani R, et al.Identifying patients with a high risk of relapse in quiescent Crohn's disease. Gut 1995; 37: 8118.
  • 35
    Mekhjian HS, Switz DM, Melnyk CS, Rankin GB, Brooks RK. Clinical features and natural history of Crohn's disease (NCCDS). Gastroenterology 1979; 77: 898906.
  • 36
    Steinhardt HJ, Loeschke K, Kasper H, Holtermuller KH, Schafer H. European Cooperative Crohn's Disease Study (ECCDS): Clinical features and natural history. Digestion 1985; 31: 97108.
  • 37
    De Domball FT, Burton IL, Clamp SE, Goligher JC. Short-term course and prognosis of Crohn's disease. Gut 1974; 15: 43543.
  • 38
    Pinchbeck BR, Imes S, Dinwoodie A, Thompson ABR. Discriminant function analysis to calculate a Crohn's activity group scale to predict future inactive or active disease. J Clin Gastroenterol 1988; 10: 498504.
  • 39
    Brignola C, Campieri M, Bazzocchi G, Farruggia P, Tragnone A, Lanfranchi GA. A laboratory index for predicting relapse in asymptomatic patients with Crohn's disease. Gastroenterology 1986; 91: 14904.
  • 40
    Farmer RG, Whelan G, Fazio VW. Long term follow up of patients with Crohn's disease. Relationship between clinical pattern and prognosis. Gastroenterology 1985; 88: 181825.
  • 41
    Wright JP. Factors influencing first relapse in patients with Crohn's disease. J Clin Gastroenterol 1992; 15: 126.
  • 42
    Siproudhis L, Mortaji A, Mary JY, Juguet F, Bretagne JF, Gosselin M. Anal lesions: any significant prognosis in Crohn's disease? Eur J Gastroenterol Hepatol 1997; 9: 23943.
  • 43
    Cottone M, Rosselli M, Orlando A, et al.Smoking habits and recurrence in Crohn's disease. Gastroenterology 1994; 106: 6438.
  • 44
    Sutherland LR, Ramcharan S, Bryant H, Fick G. Effect of cigarette smoking on recurrence of Crohn's disease. Gastroenterology 1990; 98: 11238.
  • 45
    Yamamoto T, Allan RN, Keighley MR. Smoking is a predictive factor for outcome after colectomy and ileorectal anastomosis in patients with Crohn's colitis. Br J Surg 1999; 86: 106970.DOI: 10.1046/j.1365-2168.1999.01189.x
  • 46
    Russel MG, Nieman FH, Bergers JM, Stockbrugger RW. Cigarette smoking and quality of life in patients with inflammatory bowel disease. South Limburg IBD Study Group [see comments]. Eur J Gastroenterol Hepatol 1996; 8: 107581.
  • 47
    Palli D, Trallori G, Saieva C, et al.General and cancer specific mortality of a population based cohort of patients with inflammatory bowel disease: the Florence Study. Gut 1998; 42: 1759.
  • 48
    Cosnes J, Carbonnel F, Beaugerie L, Le Quintrec Y, Gendre JP. Effects of cigarette smoking on the long-term course of Crohn's disease. Gastroenterology 1996; 110: 42431.
  • 49
    Timmer A, Sutherland LR, Martin F. Oral contraceptive use and smoking are risk factors for relapse in Crohn's disease. Gastroenterology 1998; 114: 114350.
  • 50
    Bouhnik Y, Lemann M, Mary JY, et al.Long term follow-up of patients with Crohn's disease treated with azathioprine or 6-mercaptopurine. Lancet 1996; 347: 2159.
  • 51
    Tytgat GNJ, Mulder CJJ, Brummelkamp WH. Endoscopic lesions in Crohn's disease early after ileocaecal resection. Endoscopy 1988; 20: 2602.
  • 52
    Rutgeerts P, Geboes K, Vantrappen G, Beyls J, Kerremans R, Hiele M. Predictability of the post operative course of Crohn's disease. Gastroenterology 1990; 99: 95663.
  • 53
    Rutgeerts P, Geboes K, Vantrappen G, Kerremans R, Coenegrachts JL, Coremans G. Natural history of recurrent Crohn's disease at the ileocolonic anastomosis after curative surgery. Gut 1984; 25: 66572.
  • 54
    Olaison G, Smedh K, Sjodahl R. Natural course of Crohn's disease after ileocolonic resection — endoscopically visualized ileal ulcers preceding symptoms. Gut 1992; 33: 3315.
  • 55
    DeJong E, Vandullemen HM, Slors JFM, Dekkers P, Vandeventer SJH, Tytgat GNJ. Correlation between early recurrence and reoperation after ileocolonic resection in Crohn's disease: a prospective study. J Am Coll Surg 1996; 182: 5038.
  • 56
    Lorenz-Myer H, Malchow H, Miller B, Stock H, Brandes JW. European Cooperative Crohn's disease study (ECCDS). Colonosc Dig 1985; 31: 10919.
  • 57
    Gomes P, Du Boulay C, Smith CL, Holdstock G. Relationship between disease activity indices and colonoscopic findings in patients with colonic inflammatory bowel disease. Gut 1986; 27: 925.
  • 58
    Breese EJ, Michie CA, Nicholls SW, et al.Tumour necrosis factor alpha-producing cells in the intestinal mucosa of children with inflammatory bowel disease. Gastroenterology 1994; 106: 145566.
  • 59
    Nagel E, Bartels M, Pichlmayr R. Scanning electron microscope lesions in Crohn's disease: Relevance for the interpretation of post operative recurrence. Gastroenterology 1995; 108: 37682.
  • 60
    Heinmann TM, Greenstein AJ, Lewis B, Kaufman D, Heinmann DM, Aufses AHJ. Prediction of early symptomatic recurrence after intestinal resection in Crohn's disease. Ann Surg 1993; 218: 2949.
  • 61
    Heinmann TM, Aufses AHJ. The role of peripheral lymphocytes in the prediction of recurrence in Crohn's disease. Surg Gynecol Obstet 1985; 160: 2958.
  • 62
    Markowitz J, Kahn E, Daum F. Prognostic significance of epithelioid granulomas found in rectosigmoid biopsies at the initial presentation of paediatric Crohn's disease. J Pediatr Gastroenterol Nutr 1989; 9: 1826.
  • 63
    Borley NR, Mortensen NJ, Jewell DP. Preventing postoperative recurrence of Crohn's disease. Br J Surg 1997; 84: 1493502.
  • 64
    Williams JG, Wong WD, Rothenberger DA, Goldberg SM. Recurrence of Crohn's disease after resection. Br J Surg 1991; 78: 109.
  • 65
    Musso A, Sostegni R, Astegiano M, et al.Smoking and clinical course of Crohn's disease: any adverse effect. Gastroenterology 1995; 108: A882A882(Abstract).
  • 66
    Lock MR, Farmer RG, Fazio VW, Jagelman DG, Lavery IC, Weakley FL. Recurrence and reoperation for Crohn's disease. The role of disease location in prognosis. N Engl J Med 1981; 304: 15868.
  • 67
    Greenstein AJ, Sachar DB, Pasternack BS, Janowitz HD. Reoperation and recurrence in Crohn's colitis and ileocolitis. Crude and cumulative rates. N Engl J Med 1975; 293: 68590.
  • 68
    Whelan G, Farmer RG, Fazio VW, Goormastic M. Recurrence after surgery in Crohn's disease (clinical pattern) and surgical indication. Gastroenterology 1985; 88: 182633.
  • 69
    Binder H, Freeman HJ. Risk factors predicting recurrent small intestinal resection for Crohn's disease of the terminal ileum. Can J Gastroenterol 1994; 8: 1938.
  • 70
    Holzheimer RG, Molloy RG, Wittmann DH. Postoperative complications predict recurrence of Crohn's disease. Eur J Surg 1995; 161: 12935.
  • 71
    Griffiths AM, Wesson DE, Shandling B, Corey M, Sherman PM. Factors influencing postoperative recurrence of Crohn's disease in childhood. Gut 1991; 32: 4915.
  • 72
    Basilisco G, Campanini M, Cesana B, Ranzi T, Bianchi P. Risk factors for first operation in Crohn's disease. Am J Gastroenterol 1997; 84: 74952.
  • 73
    Post S, Herfarth C, Bohm E, et al.The impact of disease pattern, surgical management, and individual surgeons on the risk for relaparotomy for recurrent Crohn's disease. Ann Surg 1996; 223: 25360.
  • 74
    Besnard M, Jaby O, Mougessot JF, et al.Postoperative outcome of Crohn's disease in 30 children. Gut 1998; 43: 6348.
  • 75
    Pallone F, Boirivant M, Stazi MA, Cosintino R, Prantera C, Torsoli A. Analysis of clinical course of postoperative recurrence in Crohn's disease of distal ileum. Dig Dis Sci 1992; 37: 2159.
  • 76
    Present DH, Meltzer SJ, Krumholz MP, et al.6-Mercaptopurine in the management of inflammatory bowel disease: short- and long-term toxicity. Ann Intern Med 1989; 111: 6419.