Endoscopy is an important diagnostic and therapeutic

Transcription

1 ORIGINAL ARTICLE Appropriateness and Diagnostic Yield of Colonoscopy in the Management of Patients with Ulcerative Colitis: A Prospective Study in an Open Access Endoscopy Service Gianpiero Manes, MD, Venerina Imbesi, MD, Sandro Ardizzone, MD, Andrea Cassinotti, MD, Matteo Bosani, MD, Alessandro Massari, MD, and Gabriele Bianchi Porro, PhD Background: Colonoscopy is frequently performed in ulcerative colitis (UC), but its benefit in the management of the disease is a matter of debate. The objective was to determine the clinical impact of colonoscopy in UC. Methods: Consecutive patients with UC undergoing colonoscopy were studied. The design and main outcome measurement was appropriateness of indications, evaluated according to guidelines. Endoscopic findings altering the management of the patients were registered. The endoscopist s management decisions based on patient s clinical picture were compared with those selected after endoscopy. Need for further investigations was recorded. Endpoints for colonoscopy-improving management were prospectively defined: change in medical therapy, need for adjuctive procedures, identification or exclusion of cancer, adenomatous polyps, or other conditions with clinical impact. The setting was an open access endoscopy service in a tertiary care center. Results: In all, 507 patients (268 male, 239 female, mean age 42 years) were included. Colonoscopy was indicated in 60.8% of cases. In 46% of patients endoscopy revealed a significant lesion; this rate was higher for indicated (67.2) than for not indicated procedures (13.5%, P ). The endoscopist s decision was altered by the endoscopic finding in 7.6% of cases and was not different between appropriate and inappropriate procedures. Conclusions: Endoscopy is a potent tool in the management of UC if correctly used. However, in the majority of cases a correct therapeutic decision may be established simply on the basis of the clinical picture. Relevant endoscopic findings have a relatively low impact on the medical treatment, but may have a very important value in the prognostic assessment of the disease. (Inflamm Bowel Dis 2008;14: ) Key Words: ulcerative colitis, colonoscopy, appropriateness, relevant finding Received for publication October 26, 2007; Accepted January 28, From the Department of Clinical Science, Chair and Department of Gastroenterology, L. Sacco University Hospital, Milano, Italy. Reprints: Dr. Gianpiero Manes, Divisione e Cattedra di Gastroenterologia, Ospedale Universitario L. Sacco, Via G.B. Grassi 74, Milano, Italy ( gimanes@tin.it). Copyright 2008 Crohn s & Colitis Foundation of America, Inc. DOI /ibd Published online 3 March 2008 in Wiley InterScience ( wiley.com). Endoscopy is an important diagnostic and therapeutic method in inflammatory bowel disease (IBD), being useful for both Crohn s disease (CD) and ulcerative colitis (UC). 1 Endoscopy is useful to make an initial diagnosis of IBD, to distinguish UC from CD, to assess the disease extent and activity, to monitor response to therapy, to allow for surveillance of dysplasia and neoplasia, and to provide endoscopic treatment, such as stricture dilation. 2,3 When used with these indications, endoscopy is likely to offer the clinician information that can be translated into an improvement of clinical management. Nevertheless, in spite of its wide use, no study has evaluated the real impact of endoscopy in the management of UC. The potential importance of this issue is evident, in that colonoscopy is an invasive and costly procedure. Colon cleansing and the procedure itself may be unpleasant especially for individuals with active IBD. Overuse and inappropriate use of colonoscopy could result in an unnecessary discomfort for patients with UC; on the other hand, a correct use of colonoscopy could determine a substantial reduction of the endoscopic workload and better channel the endoscopic resources into necessary care. From a theoretical point of view an examination can be regarded as useful when able to alter the management strategy of a disease in such a way that is not predictable before performing the procedure. Regarding the use of colonoscopy in UC patients, a number of questions still remain unanswered. What is the clinical benefit of performing a colonoscopy in different clinical situations? How often does a colonoscopy improve the management of IBD? Which subgroups of IBD patients can benefit the most from a colonoscopy? The aim of the present study was to determine the impact of utilizing colonoscopy on the management of UC in the wide spectrum of clinical presentations that gastroenterologists may face in their everyday activity. MATERIALS AND METHODS The study was conducted in a 2-year period (February 2005 to January 2007) in a single endoscopy service in Milan, a third-level center where about 3500 patients with IBD are cared for. All UC patients who presented during the study period to perform a colonoscopy were prospectively consid- Inflamm Bowel Dis Volume 14, Number 8, August

2 Manes et al Inflamm Bowel Dis Volume 14, Number 8, August 2008 ered if they were older than 18 years and gave informed consent to participate in the study. The study was approved by the local ethics committee. Patients were referred to endoscopy by a team of 4 physicians: 2 are expert gastroenterologists with experience in IBD who supervised 2 younger specialists. Prior to endoscopy, patients were interviewed by the referring gastroenterologist and the following data were recorded in a structured form: age, sex, symptoms (characteristics, intensity, and duration), extent of colonic involvement, time of the first UC diagnosis, previous and current treatment with regard to the use 5-aminosalycilic acid (5-ASA), steroids, and immunosuppressors, previous colonic surgery, previous endoscopic and radiologic examinations, and concomitant diseases. A symptom score was also calculated for each patient according to the modified Powell Tuck index. 4 During endoscopy, the severity of colitis, graded according to Baron s score, 5 and the extent of disease were recorded by the endoscopists (G.M. and V.I.). The presence of vascular lesions, polyps, neoplasia, strictures, and other significant lesions were also registered. Colonoscopy features were analyzed with respect to: appropriateness of colonoscopy; a significant diagnosis, i.e., the presence of a relevant finding; and colonoscopy-based improvement of management. Appropriateness of Colonoscopy The appropriateness of the indications for endoscopy was judged by the endoscopist immediately before the procedure after having interviewed the patient and considered the clinical picture of the disease. Appropriateness was deduced by a critical analysis of the current literature and ASGE guidelines. 1 In particular, endoscopy was considered indicated if performed: 1. To make the initial diagnosis of IBD and to distinguish between UC and CD. 1,6,7 2. To evaluate the disease extent and activity in uninvestigated patients. 1,6,7 3. To assess the response to steroids therapy in patients with acute disease. 8,9 4. To investigate refractoriness to therapy, 1 i.e., lack of response to a therapy after a stable clinical remission induced by the therapy itself. 5. To investigate patients with severe acute disease In case of symptoms suggesting complications or a malignancy As a screening for colorectal cancer (CRC) after 10 years of disease in patients with pancolitis or 15 years in the case of left-sided colitis. 1,11 8. To monitor ileal pouch anal anastomosis. 1,12 Endoscopy was considered not indicated if performed: i. In patients with stable disease. ii. To evaluate patients with mild to moderate recurrence of disease. 13 Relevant Findings of Colonoscopy The analysis was aimed at examining the discriminant variable of having a relevant diagnosis revealed by a colonoscopy. A diagnosis was considered relevant if able to improve the management decision. This happened when any of the following scenarios occurred: a. The medical therapy of UC was altered based on the endoscopic picture. b. The endoscopic finding required further testing or operative procedures not otherwise hypothesized before endoscopy. c. Endoscopy offered new findings which modified the management of the disease: a change in the initial diagnosis or in the previously reported extension of the disease; a different disease activity than otherwise hypothesized; the presence of complications such as strictures. d. A diagnosis of CRC, adenomatous polyp, or dysplasia. The relevance of endoscopy was evaluated by 2 authors (S.A. and A.C.) after having accurately analyzed the patient s chart, symptoms, and clinical history. Colonoscopy-based Improvement of Management Prior to perform colonoscopy, the referring gastroenterologist was requested to select a therapeutic decision based on a patient s presenting symptoms and clinical history: increase, maintain, or decrease the current treatment. Adding a new medication, increasing the dosages of the current treatment, or switching from a class to another class of more potent medications (i.e., 5-ASA to steroids, steroids to immunosuppressors) were all considered as an increase in the treatment. The hypothesized need for further investigations and operative procedures, dilation, surgery, etc., was also registered. After colonoscopy, the gastroenterologist was asked again to select a therapeutic decision considering both the patient s symptoms and the endoscopic findings: increase, maintain, or decrease the current therapy. The need for further testing and operative procedures was also recorded. To maintain blindness, the 2 endoscopists were unaware of the gastroenterologist s findings when performing colonoscopy. Data Analysis Analysis was aimed at examining the determinants of 3 dependent variables: 1) appropriateness of colonoscopy; 2) having a significant diagnosis revealed by colonoscopy; 2) having a colonoscopy that changes the management decision. Descriptive statistics consisted of t-tests for continuous variable and 2 tests for categorical variables. Univariate logistic 1134

3 Inflamm Bowel Dis Volume 14, Number 8, August 2008 Colonoscopy in Ulcerative Colitis TABLE 1. Demographic and Clinical Characteristics of the Study Patients All Patients Diagnosis- Staging Posttreatment Refractory Disease Stable Disease Severe Acute Disease Symptom Relapse CRC Screening Others Number (%) 507 (100) 35 (6.9) 115 (22.7) 33 (6.5) 123 (24.3) 20 (3.9) 76 (15) 80 (15.8) 25 (4.9) Gender (M/F) 268/239 16/19 67/48 13/20 59/64 12/8 44/32 44/37 13/12 Age (mean, range) 42.3 (18 67) 35 (18 53) 32 (23 55) 39.6 (29 49) 46.3 (30 56) 41.2 (23 52) 47.5 (29 58) 51.9 (36 67) 38.6 (23 45) Disease duration, yrs (SD) 7.9 (4.9) 0.7 (1.5) 0.8 (1.6) 6 (4.7) 9.6 (6.5) 7.4 (5.7) 8.3 (6.7) 19.4 (6.7) 7.7 (6.4) Disease extension (%) Pouch 12 (2.4) (48) Rectum 69 (13.6) 8 (22.8) 14 (12.2) 3 (9.1) 36 (29.3) 4 (20) 4 (5.3) 0 0 Left colitis 214 (42.2) 13 (37.2) 58 (50.4) 22 (66.7) 66 (53.6) 8 (40) 37 (48.7) 10 (12.5) 0 Pancolitis 212 (41.8) 14 (40) 43 (37.4) 8 (24.2) 21 (17.1) 8 (40) 35 (46) 70 (97.5) 13 (52) Current treatment (%) No treatment 92 (18.2) 11 (31.4) (25.2) 5 (25) 20 (26.3) 15 (18.7) 10 (40) 5-ASA 211 (41.6) 13 (37.2) 0 26 (78.8) 65 (52.8) 8 (40) 37 (48.7) 54 (67.5) 8 (32) Steroids 157 (30.9) 11 (31.4) 115 (100) 7 (21.2) 5 (4.1) 3 (15) 9 (11.8) 5 (6.3) 2 (8) Immunosuppressors 47 (9.3) (17.9) 4 (20) 10 (13.2) 6 (7.5) 5 (20) Symptoms score (%) No symptoms 223 (44) 0 59 (51.3) 0 93 (75.6) (86.2) 2 (8) Mild 116 (22.9) 23 (65.8) 9 (7.8) 3 (9.1) 30 (24.4) 0 25 (32.9) 11 (13.8) 15 (60) Moderate 110 (21.7) 6 (17.1) 30 (26.1) 24 (72.7) (59.2) 0 5 (20) Severe 58 (11.4) 6 (17.1) 17 (14.8) 6 (18.2) 0 20 (100) 6 (7.9) 0 3 (12) Referral clinic (%) Outpatient 333 (65.7) 6 (17.1) 108 (93.9) 6 (18.2) 96 (78) 4 (20) 22 (29) 80 (100) 11 (44) Inpatient 174 (34.3) 29 (82.9) 7 (6.1) 27 (81.8) 27 (22) 16 (80) 54 (71) 0 (0) 14 (56) Patients are grouped according to the indication to endoscopy (see also the text). The group other includes 12 patients who underwent endoscopy to follow-up an ileal pouch anal anastomosis, 8 patients who required endoscopy for the occurrence of obstruction symptoms, and 5 for the presence of alarm symptoms (2 weight loss, 2 anemia, 1 abdominal mass). The symptom scores were calculated according to the Powell-Tuck index and grouped as following: 3-5 mild, 6-8 moderate, 9 severe. regression was used to test the significance of the characteristics with dependent variables. Multivariate logistic regression was conducted to evaluate the association among the determinants with a dependent variable while simultaneously controlling for the effect of other variables. A Pearson s analysis was performed to evaluate the correlation between the symptom score and the severity of mucosal lesions. Statistical significance was defined as P RESULTS A total of 1140 patients with IBD, 580 of them with UC, underwent endoscopy in the study period. In all, 507 met the study criteria and agreed to participate. Of these, 333 patients (65.7%) were referred to endoscopy by the outpatient clinic and 174 (34.3%) by the inpatient clinic. The demographic and clinical characteristics of the study patients according to the indications for endoscopy are reported in Table 1. Appropriateness of Colonoscopy in UC Patients Indications for colonoscopy in patients with UC are reported in Table 1. According to the current literature, a colonoscopy indication was considered not appropriate in 199/507 patients (39.2%). Inappropriate endoscopy was performed in patients with stable disease on remission (123, 24.2%) and in patients with a mild/moderate clinical relapse (76, 15%). Relevant Findings of Colonoscopy Severity of mucosal lesions as assessed by means of Baron s score correlated significantly with the symptoms score (r 0.784, P 0.01). The relevant endoscopic findings observed at colonoscopy according to the indications for the procedure are reported in Table 2. Overall, 46% of endoscopies revealed a significant lesion. The rate of relevant findings was significantly higher for the indicated than for the not indicated procedures (67.2% versus 13.5%, P ), while it was not different with regard to gender, age, and referral source. The likelihood of obtaining a relevant finding was 5.8 times (95% confidence interval [CI] ) higher for an appropriate endoscopy than for an inappropriate procedure. In particular, endoscopy performed to obtain the first diagnosis 1135

4 Manes et al Inflamm Bowel Dis Volume 14, Number 8, August 2008 TABLE 2. Relevant Findings Achieved by Colonoscopy with Respect to the Indications to the Procedure Relevant Findings (%) Type of Findings (number) All patients 234/507 (46.1) Diagnosis/staging 35/35 (100) First diagnosis of UC (18) Extension of disease (17) Posttreatment 35/115 (30.4) Severe colitis (35) Refractory disease 18/33 (54.5) More extended colitis (6) Deep ulcers (9) Cytomegalovirus infection (1) Stable disease 18/123 (14.6) More extended colitis (3) Moderate activity (15) Severe acute disease 14/20 (70) Deep ulcers (3) Severe lesions (11) Symptom relapse 9/76 (11.8) Deep ulcers (3) Severe lesions (6) CRC screening 80/80 (100) Absence of cancer and dysplasia (80) Others 25/25 (100) Pouchitis (12) Strictures (3) Polyps (2) Exclusion of complication (8) or to stage the disease in naïve patients achieved, as expected, 100% of relevant findings. A rate of 100% was also observed when endoscopy was performed for the suspicion of complications, pouchitis, or malignancies and in the screening for CRC. Pouchitis, polyps, and stenosis were the relevant findings in these groups of patients, but also a normal finding was considered relevant when it excluded complications or malignancies. The rate of relevant findings was lower for endoscopies performed with other indications. Relevant findings achieved in patients with severe acute disease and after a steroid treatment of an acute episode were mainly severe colitis and deep ulcers; these lesions were considered relevant since they are likely to have a prognostic value. A more extended colitis than expected, presence of deep ulcers, and demonstration of cytomegalovirus infection were considered relevant findings in patients who did not respond to the treatment since they are likely to be the cause of refractoriness (Table 2). In patients with stable disease and with mild to moderate recurrence the few relevant findings were mainly the presence either of a more severe or of a more extended mucosal damage than expected on the basis of the symptoms. These lesions were significant since they were able to influence the further management of the disease (Table 2). Colonoscopy-based Improvement of Management The gastroenterologist s decision regarding medical therapy is provided in Table 3. A high concordance rate (92.4%) was observed between what the gastroenterologist would have done before and after endoscopy. As a consequence, the prevalence of a colonoscopy-based improvement of management decision was quite low, not exceeding 7.6%, and was not significantly different with regard to gender, age, and referral source. Interestingly, the rate of endoscopies with a significant impact on management was not different between appropriate and inappropriate examinations (22/272, 8.1% versus 21/199, 10.6%), suggesting that the short-term management decision of patients with UC should be based mainly on the clinical evaluation. Endoscopies performed in naïve patients to make a first diagnosis of UC and to evaluate TABLE 3. Gastroenterologist s Decision Regarding Medical Therapy Before and After Colonoscopy Increase (%) Maintain (%) Decrease (%) Before After Before After Before After Concordance (%) All patients 197 (41.8) 224 (47.5) 264 (55.9) 236 (50) 11 (2.3) 12 (2.5) 428/472 (92.4) Diagnosis-staging (35) 30 (85.7) 5 (4.3) 0 (0) Not considered Posttreatment (115) 47 (40.8) 55 (47.8) 68 (59.2) 60 (52.2) 0 (0) 0 (0) 107/115 (93.4) Refractory disease (33) 33 (100) 33 (100) 0 (0) 0 (0) 0 (0) 0 (0) 33/33 (100) Stable disease (123) 2 (1.6) 17 (13.8) 118 (95.9) 103 (83.7) 3 (2.5) 3 (2.5) 108/123 (87.8) Severe acute disease (20) 20 (100) 20 (100) 0 (0) 0 (0) 0 (0) 0 (0) 20/20 (100) Symptom relapse (76) 70 (92.1) 76 (100) 6 (7.9) 0 (0) 0 (0) 0 (0) 70/76 (92.1) CRC screening (80) 0 (0) 6 (7.6) 72 (90) 65 (81.2) 8 (10) 9 (11.2) 73/80 (91.2) Others (25) 25 (100) 17 (68) 0 (0) 8 (32) 0 (0) 0 (0) 17/25 (68) Endoscopies performed in naïve patients to make the first diagnosis of IBD and to evaluate the disease extent and activity were excluded from this analysis since the gastroenterologists did not have enough information to express their management decision before endoscopy. 1136

5 Inflamm Bowel Dis Volume 14, Number 8, August 2008 Colonoscopy in Ulcerative Colitis the disease extent and activity were excluded from this analysis since the gastroenterologists did not have enough information to express their management decision prior to endoscopy. Endoscopy changed the clinical management mainly when the severity of mucosal lesions was different than expected on the basis of the symptoms. In 5 patients with a suspicion of an intestinal obstruction and in 3 with alarm symptoms, for whom the physician had hypothesized performing further investigations or surgery, endoscopy altered the management excluding the suspected conditions. DISCUSSION In spite of the recognized value of colonoscopy, no study has evaluated how endoscopy may practically influence the management of UC. From a theoretical point of view, endoscopy may be considered useful if able to detect relevant lesions, i.e., lesions that have direct therapeutic and/or prognostic consequences, a new diagnosis of UC, the definition of disease activity, or the diagnosis of a complication. However, the definition of relevant finding may be nebulous and the problem is even bigger when we try to define what is irrelevant. The endoscopic diagnosis of an active colitis may be an irrelevant finding in patients with a relapse of UC, but may be very relevant if the patients have few symptoms. In the present study we introduced the concept of colonoscopy-based improvement of management and, accordingly, we defined useful an endoscopy that is able to change the management decision in a way that is not predictable prior to performing the investigation. This concept partly resembles that of diagnostic yield and parallels that of relevant finding, but while the diagnostic yield of a method is usually evaluated a priori according to what has been previously defined useful in the management of the disease, in the present study usefulness of colonoscopy was evaluated a posteriori, comparing what the endoscopist would have done before endoscopy and what he/she will do after endoscopy. One important finding arising from our study is that in the majority of cases a correct, short-term therapeutic decision may be established simply on the basis of the symptoms and, thus, without performing endoscopy; conversely, most of the relevant lesions demonstrated by endoscopy have mainly a prognostic value and are relevant especially in the long-term management of the patients. This is the case for 2 somewhat controversial recommendations for endoscopic testing: the occurrence of a severe attacks of UC and the evaluation of mucosal healing after steroid treatment in patients with acute colitis. In the first case endoscopy is usually prescribed to evaluate the presence of endoscopic signs of a poor prognosis, severe mucosal lesions and deep ulcerations. An endoscopically severe colitis was found in our series in 11/20 (55%) of patients, and deep ulcerations in 3 (15%). Medical therapy was not altered by endoscopy in any of these patients since an increase in the doses or a switching to a more potent drug were established in all of them before endoscopy on the basis of the symptoms. Some studies suggest that the presence of severe lesions is correlated with a lower rate of response to medical therapy (7% versus 74% in patients with severe versus moderate colitis, respectively) and a 41 times higher risk of undergoing colectomy. 10,14 The relevance of this information would have probably been evident in the further follow-up of our patients. With regard to the second point, that is, endoscopy performed after a medical treatment for acute colitis, there is some evidence that the endoscopic remission of colitis likely predicts a more sustained clinical remission and a better long-term outcome in terms of colectomy rate, immunosuppressant use, and relapse rate. 8,9 In our study, 68/115 (59%) of patients investigated with this indication became either asymptomatic or improved their symptoms significantly under steroid treatment, but in only 36 of these patients did the mucosal lesions heal and in 8 of them severe lesions were still present. For all these patients the risk of an early recurrence is likely to be high 8,9 and, at least from the theoretical point of view, medical therapy should be prolonged or increased. Our study was not designed to evaluate this aspect since a postendoscopy follow-up was lacking; however, the likelihood of undertreating these patients is high if we do not perform an endoscopic check at the end of the medical treatment. Some indications to colonoscopy are considered appropriate by most authors dealing with IBD: to make the initial diagnosis of UC, to distinguish CD from UC, to assess the disease extent and activity in uninvestigated patients, and in the screening of CRC in patients with long-lasting disease. 1 In our study we confirmed that endoscopies performed with these indications are undoubtedly useful and offer valuable information to the physician. In the case of endoscopies performed in the prevention of CRC, no cancer or dysplasia were found in our series. The negative finding was, nevertheless, considered a relevant finding since it was able to exclude a pathology, and thus with a significant impact on the management of disease. Similar is the case of endoscopies performed for the suspicion of complications or malignancies. Either positive or negative findings should be considered significant and are able to alter the further management of the patients. Another appropriate indication is the occurrence of refractoriness to therapy. In the majority of these patients endoscopy was apparently not able to alter the management of the disease since an increase in treatment was already predictable prior to endoscopy. In 50% of patients, however, endoscopy offered to the physician information on the putative mechanisms at the basis of refractoriness: a more extended colitis than expected in 6 patients on treatment with 1137

6 Manes et al Inflamm Bowel Dis Volume 14, Number 8, August 2008 topical 5-ASA, a very severe disease with deep ulcers in 9 patients, and a cytomegalovirus infection in 1 patient. About 40% of endoscopies performed in our study were judged to be not indicated according to the current literature, for instance, in patients with stable disease and with mildmoderate recurrence of symptoms. This apparently high rate of inappropriate procedures is the consequence of the fact that ours is a tertiary reference center for IBD where patients are often included in clinical studies and submitted to a very strict follow-up. We do not know whether our series may be representative for other high-volume IBD-centers, but, to our knowledge, this is the first study specifically designed to evaluate the appropriate use of colonoscopy in patients with UC. Since a discrepancy may exist between symptoms and mucosal appearance and symptoms do not always reflect reliably disease severity, 10,15 endoscopies not indicated could potentially be of practical clinical importance. This would mean, at least from the theoretical point of view, that patients with stable disease may have severe endoscopic lesions and, on the other hand, a symptomatic relapse would not necessarily be associated with an endoscopic relapse. In our study good correlation was found between the severity of symptoms and the degree of mucosal lesions; the consequence was that the number of relevant findings detected by inappropriate colonoscopies was very low and significantly lower than by appropriate procedures. Moreover, in only 12% of patients did inappropriate endoscopies change the management of UC. Our study thus confirms that patients with stable disease should not undergo an endoscopic follow-up (if not to prevent CRC) and that in case of a clinical recurrence of disease a change in the treatment can be established simply on the basis of symptoms. This is what usually happens in clinical practice and a recent study has also demonstrated that endoscopic findings provide little additional information for the clinical indices of disease activity. 13 Our study design deserves some considerations. We defined a scenario on how endoscopy could improve the management of UC, but we have not included a follow-up to evaluate the impact of treatments and of different endoscopic findings on the clinical course of disease. Moreover, the gastroenterologists referring the patients to endoscopy were not blinded to the study hypotheses, and this could represent a degree of bias. We will probably need further studies with a proper design to consider these 2 aspects of the problem. In summary, endoscopy is a very potent tool in the management of patients with UC, provided that it is used in the correct way. With the exception of naïve patients where endoscopy is fundamental to diagnose and stage the disease, short-term management of UC can be established in the majority of cases simply on the basis of the clinical picture. Nevertheless, endoscopy may offer the physician important data which are likely to have a relatively low impact on the medical treatment, but are of great value in the management of disease in terms of prognosis and, thus, in the long-term management decision. REFERENCES 1. Anonymous. ASGE guideline: endoscopy in the diagnosis and treatment of inflammatory bowel disease. Gastrointest Endosc. 2006;63: Chutkan RK, Scherl E, Waye JD. Colonoscopy in inflammatory bowel disease. Gastrointest Endosc Clin North Am. 2002;12: Hommes DW, van Deventer SJ. Endoscopy in IBD. Gastroenterology. 2004;126: Powell-Tuck J, Day DW, Buckell NA, et al. Correlation between defined sigmoidoscopic appearance and other measures of disease activity in ulcerative colitis. Dig Dis Sci. 1982;27: Baron JH, Connel AM, Lennard-Jones JE. Variation between observers in describing mucosal appearances in proctocolitis. Br Med J. 1964; 5373: Pera A, Bellando P, Caldera D, et al. Colonoscopy in inflammatory bowel disease. Diagnostic accuracy and proposal of an endoscopic score. Gastroenterology. 1987;92: Abreu MT, Harpaz N. Diagnosis of colitis: making the initial diagnosis. Clin Gastroenterol Hepatol. 2007;5: Courtney MG, Nunes DP, Bergin CF, et al. Colonoscopic appearance in remission predicts relapse of ulcerative colitis. Gastroenterology. 1991; 100S:A Ardizzone S, Cassinotti A, Penati MC, et al. Clinical and endoscopic outcome after the first corticosteroid course in newly diagnosed ulcerative colitis: a 5-year follow-up inception cohort study. Gastroenterology. 2007;132S:A Carbonnel F, Lavergne A, Leheman M, et al. Colonoscopy of acute colitis. A safe and reliable tool for assessment of severity. Dig Dis Sci. 1994;39: Eaden JA, Mayberry JF. British Society for Gastroenterology, Association of Coloproctology for Great Britain and Ireland. Guidelines for screening and surveillance of asymptomatic colorectal cancer in patients with inflammatory bowel disease. Gut. 2002;51(S5):V Sandborn WJ, Tremaine WJ, Batts KP, et al. Pouchitis after ileal pouch-anal anstomosis: a Pouchitis Disease Activity Index. Mayo Clin Proc. 1994;69: Higgins PDR, Schwartz M, Mapili J, et al. Is endoscopy necessary for the measurement of disease activity in ulcerative colitis? Am J Gastroenterol. 2005;100: Daperno M, Sostegni R, Scaglione N, et al. Outcome of a conservative approach in severe ulcerative colitis. Dig Liver Dis. 2004;36: Floren CH, Benoni C, Willen R. Histologic and colonoscopic assessment of disease extension in ulcerative colitis. Scand J Gastroenterol. 1987; 22: