Trials in Prevention of Post Surgical Recurrence in Crohn s Disease

Transcription

1 Send Orders of Reprints at Reviews on Recent Clinical Trials, 2012, 7, Trials in Prevention of Post Surgical Recurrence in Crohn s Disease Claudio Papi*, Federica Fascì Spurio, Giovanna Margagnoni and Annalisa Aratari Gastroenterology & Hepatology Unit, S Filippo Neri Hospital, Via Martinotti Rome, Italy Abstract: Most patients with Crohn's disease will require surgery during the course of their disease. However, surgery is not curative and post-operative is quite inexorable. One year after resection up to 80% of patients have new lesions at the neo-terminal ileum and after 10 years approximately 50% of patients will experience of symptoms and 35% will need further surgery. Prevention of post-operative has, therefore, a central role in the management of Crohn's Disease. Several drugs have been evaluated to decrease the risk of both endoscopic and clinical but the overall results are largely not impressive. Among the different drugs evaluated, mesalazine, antibiotics (metronidazole and ornidazole), thiopurines and anti-tnf antibodies have been shown to be effective whereas budesonide, probiotics and interleukin 10 are not effective. This review focuses on the actual evidence on the prevention of postoperative : randomìsed controlled trials and meta-analyses are critically reviewed and discussed with particular attention to the methodological aspects. Keywords:????????????????????. Surgical resection is an almost inevitable event through the course of Crohn s disease (CD) and the surgical rate increases with time. The probability of surgery is 20-40% during the first year after diagnosis and ranges between 30% and 70% 10 years after diagnosis; after 15 years, the probability of surgery is as high as 70-90% [1,2]. Surgery is not curative and post-operative (POR) follows a predictable and sequential course. One year after resection, approximately 60 80% of patients have new lesions at the neoterminal ileum (endoscopic ), 10 20% will develop symptoms (clinical ) and 5% will need further intestinal resection (surgical ) [3]. After 10 years, approximately 50% of patients will experience clinical and up to 35% will need re-operation [4,5]. Several prognostic factors have been investigated as potentially influencing POR. There is consistent evidence that smoking increases the risk of clinical and surgical [6 8]. Penetrating behaviour is also associated with a higher risk of in most studies [9]. Other factors, including age at disease onset, gender, family history of CD, duration of disease before surgery, previous resections, length of resected bowel along with anastomotic technique, presence of granuloma in the surgical specimen, disease involvement at the line of resection, need of blood transfusions and postoperative complications have been investigated but for most of these factors the evidence is largely inconclusive [10]. Severity of endoscopic lesions in neo-terminal ileum one year after surgery is the strongest predictor of the subsequent clinical course [3]. According to the Rutgeerts score, patients with no lesions (i0) or only mild lesions (i1, less than 5 * Address correspondence to these author at the Gastroenterology and Hepatology Unit, San Filippo Neri Hospital, Via Martinotti 20, Roma, Italy; Tel: ; Fax: ; s: c.papi@fastwebnet.it; c.papi@sanfilipponeri.roma.it aphtous lesions) have a good outcome; conversely, patients with severe endoscopic (i2, more than 5 aphtous lesions; i3, diffuse ileitis; i4, large ulcers or narrowing) have a high probability of developing symptoms in the subsequent years. Prevention of post-operative is a central problem in the management of CD. Different drugs have been investigated in several randomised controlled trials (RCTs) and many meta-analyses have been performed (Table 1). When interpreting the results of RCTs and meta-analyses some aspects should be considered: 1) the definition of the outcome (endoscopic, clinical or surgical ), and 2) the estimation of rate among patients receiving placebo. DEFINITION OF THE OUTCOME Surgical is a clinically relevant end point. Repeated surgeries in CD increase the risk of short bowel syndrome, affect quality of life and lead to disability. Nevertheless, no RCT has considered surgical as a primary or secondary end point. The cumulative probability of reoperation is approximately 30% within 10 years [1] but is rare in the few years after the first resection. Therefore, to obtain reliable data, a large number of patients and a long follow up would be required when designing a RCT having surgical as a primary end-point. is usually the primary outcome in most RCTs. However, demonstration of a reduction in the risk of clinical requires an adequate sample size because the baseline risk within 1 to 2 years after surgery is relatively low [11]. For this reason endoscopic if often used as a surrogate end point considering that the severity of endoscopic lesions is related to the subsequent clinical outcome [3]. In this setting only severe endoscopic (usually defined as Rutgeerts score > 2 or > 3) appears to be a /12 $ Bentham Science Publishers

2 2 Reviews on Recent Clinical Trials, 2012, Vol. 7, No. 4 Papi et al. Table 1. Published RCTs and Meta Analyses of Different Drugs Addressing Prevention of post-operative Recurrence in Crohn s Disease RCTs Meta- Analyses Mesalazine 8 4 Antibiotics 2 1 Probiotics/Symbiotics 5 2 Budesonide 2 4 Thiopurines 6 2 IL-10 1 Infliximab 1 valid surrogate end point: in fact patients with mild lesions or no endoscopic at one year have a similar risk of clinical in the subsequent years [3]. IN CALCULATING SAMPLE SIZE, WE ESTIMATED THE SEVERE RECURRENCE RATE IN THE PLA- CEBO GROUP An accurate estimation of the rate of clinical and endoscopic in CD patients operated on for resection and receiving placebo is crucial for interpreting results of RCTs, for assessing the effect size and for calculating sample size. A meta-analysis of the placebo rates of clinical and severe endoscopic in postoperative CD showed a pooled estimate of 23.7% (95% CI 13% - 35%) and 50.2% (95% CI, 28% 73%) respectively [12]. However, significant heterogeneity among studies was found with a rate of clinical ranging from 0% to 78% and a rate of severe endoscopic ranging from 30% to 79%. This heterogeneity may reflect different patient population, and interobserver variability in the evaluation of [12]. Most RCTs evaluating the efficacy of different drugs for preventing POR consider both endoscopic and clinical as end points. This can lead to problems in calculating the sample size of the study. For example, if severe endoscopic is considered the primary end point and the sample size is calculated assuming a baseline risk of severe endoscopic of approximately 50% as reported in the meta-analysis [12], the same study may be underpowered to detect a statistically significant difference in the clinical considering that the baseline risk of clinical is lower, approximately 20% [12]. MESALAZINE Mesalazine has been extensively evaluated for the prevention of POR in CD after surgery. From 1994 to 2004 eight RCTs have been published comparing mesalazine to no treatment [13], placebo [14-18], or placebo and mercaptopurine in a three arms RCT [19]. Finally, one RCT compared two different doses of mesalazine [20]. The duration of treatment ranged from 3 to 72 months with a mesalazine dose ranging from 2.4g/daily to 4.0 g/daily. With the exception of the study of Florent et al [16], that considered only early endoscopic (12 weeks after surgery) as primary end point, all RCTs considered both endoscopic and clinical. In most studies clinical was defined according to Crohn s Disease Activity Index (CDAI) but the definition of was different across studies: CDAI > 150 and increase of at least 100 points [13,15,20]; CDAI > 150 and increase of at least 60 points [17]; CDAI > 250 or CDAI > 200 and an increase of at least 60 points [18]. Moreover, some studies defined clinical according to investigators judgment [14,17] or when complications occurred [18]. Finally, in the study of Hanauer et al [19], the CDAI was not calculated and clinical was defined by a subjective, unvalidated score system (mild, moderate or severe symptoms). was considered as an end point in all studies except in the Sutherland et al. study [17] and the Rutgeerts score was consistently used to classify endoscopic. None of the individual studies showed a significant effect of mesalazine for preventing POR with the exception of the study of Caprilli et al [13] that, however, has some methodological limitations such as the unblinded design and the lack of a placebo group. The results of the individual studies have been pooled in several meta-analyses [21-26] which differ in regards to criteria for considering studies for the review, interpretation in data extraction, statistical analysis, definition of the outcome (clinical, endoscopic or both), and measure of the treatment effect size. The characteristics of published meta-analyses are shown in Table 2. Despite all methological differences, the results are similar: mesalazine appears to be more effective than placebo for preventing clinical but the benefit is small, with a risk difference of approximately 10% and a number needed to treat (NNT) ranging from 7 to 12 (Table 3). A NNT ranging from 16 to 19 is reported in the meta-analysis of Gordon et al [25], in which the NNT was calculated using the minimum and maximum control risk of relapse among the included studies. As far as concern severe endoscopic, this end point has been considered in 2 meta-analyses [23,24] reporting different results: a risk difference of -22% (-35% to - 10%), NNT=5 [23] and a risk ratio of 0.50 ( ), NNT = 8 [24]. In conclusion, mesalazine appears to be more effective than placebo for preventing clinical and severe endoscopic in CD although the absolute benefit is small. Moreover there is a significant heterogeneity across individual studies in the time points for reporting both clinical and endoscopic, therefore the long term impact of prophylaxis is not clear. In

3 Trials in Prevention of Post Surgical Recurrence in Crohn s Disease Reviews on Recent Clinical Trials, 2012, Vol. 7, No. 4 3 Table 2. Meta Analyses Addressing Mesalazine for Prevention of Post-Operative Recurrence in Crohn s Disease Camma 1997 Cottone 2000 (update) Camma 2002 (update) Doherty 2009 Gordon 2011* Ford 2011 * Studies included Caprilli 2003 Korelitz 1998 Hanauer 2004 Hanauer 2004 Hanauer 2004 Florent 1996 Florent 1996 End points considered (overall) (overall) Clinical (severe) (severe) Outcome Risk Difference Risk Difference Risk Difference Risk Ratio Odds Ratio Relative Risk Measure (RD) (RD) (RD) (RR) (OR) (RR) * The meta-analysis included RCTs in which Sulphasalazine was used. In the table only subgroup analysis of RCTs using mesalazine is reported Table 3. Results of Meta Analyses Addressing Mesalazine for Prevention of Post-Operative Recurrence in Crohn s Disease Camma 1997 Cottone 2000 (Update) Camma 2002 (Update) Doherty 2009 Gordon 2011* Ford 2011 * Clinical RD -13.1% (95%CI -21.8% to - 4.5%) RD -10% (95%CI -16.9% to - 3.2%) RD -15% (95%CI -23% to - 8%) RR 0.76 (95%CI ) OR 0.69 (95%CI ) RR 0.80 (95%CI ) NNT = 8 NNT = 10 NNT = 7 NNT=12 NNT = 16-19** NNT = 10 RD -18% Overall Recurrence (95%CI -24% to - 10%) NNT = 6 RR 0.93 (95%CI ) Severe Recurrence RD -22% (95%CI -35% to - 10%) RR 0.50 (95%CI ) NNT = 5 NNT = 8 * The meta-analysis included RCTs in which Sulphasalazine was used. In the table only subgroup analysis of RCTs using mesalazine is reported ** The NNT was calculated using the minimum and maximum control risk of relapse amongst the included studies a retrospective observational study of 216 patients with ileocaecal CD at their first resection and with a long postoperative follow up (mean months, range ), the cumulative probability of clinical and surgical within 10 years was not different in patients receiving and not receiving regular mesalazine prophylaxis [27].

4 4 Reviews on Recent Clinical Trials, 2012, Vol. 7, No. 4 Papi et al. Poor adherence to the treatment could be one explanation of the negative results of this long-term observational study. It is well known that poor adherence to maintenance treatment of inflammatory bowel disease (IBD) is frequently observed. However, in this observational study, the problem of poor adherence to the treatment could be minimized by the fact that patients included in the study were followed at regular intervals at two IBD referral centres were the compliance to treatment is regularly assessed. ANTIBIOTICS The etiology of IBD is unknown but may relate to an unidentified bacterial pathogen or an immunological reaction to gut microbiota. A large body of evidence from animal models and clinical observations suggests that luminal bacteria are the most probable inducer of chronic inflammation in CD. Therefore antibiotics have been proposed as a primary therapy for CD but current data are conflicting. Two studies from the same author evaluated the efficacy of nitroimidazole antibiotics (metronidazole and ornidazole) for prevention of POR [28,29]. In the first study [28], 60 patients who underwent curative ileal resection and primary anastomosis were randomised to receive metronidazole 20 mg/kg or placebo for 3 months. The main outcome was early endoscopic, defined as a Rutgeerts score of 3 or 4, three months after surgery. The sample size was calculated assuming an early rate of 70% in the placebo group. On per-protocol analysis metronidazole reduced the risk of severe endoscopic by 30% (43% in the placebo group, 13% in the metronidazole group) and the difference reached statistical significance (P = 0.02). However, on intention to treat analysis, there was not significant difference between metronidazole and placebo due to the high rate of dropouts because of metronidazole side effects. In the subsequent 3 years follow-up, the cumulative probability of a course free of symptoms, based on intention-totreat analysis, was not different in the active treatment and placebo groups. In a subsequent trial [29], 80 patients were randomized to ornidazole 1 g/daily or placebo started within 1 week after resection and continued for 1 year. On per-protocol analysis, ornidazole reduced severe endoscopic at 3 and 12 months by 24.4% (95% CI, 4% 45.7%) and 25.2% (95% CI, 1.6% 45%) respectively. On intention to treat analysis, clinical at 1 year (the primary end point) occurred in 7.9% of patients in the ornidazole group and in 37.5% in the placebo group (risk difference 29.6%; 95% CI, 11.5% 39.4%). Overall nitro-imidazole antibiotics are effective for reducing both clinical and endoscopic, but appear to be poorly tolerated and their use as a long-term treatment is limited by adverse effects. In a meta-analysis of the two RCTs, patient withdrawal and adverse events were both significantly greater with nitroimidazole therapy compared to placebo [24]. The question if a short term antibiotic course may have a starter effect on flora and can induce a positive long-term effect on disease course is of great interest. There are no data in the literature exploring specifically this concept. PROBIOTICS Five studies (one published as an abstract only) evaluated the effect of different probiotic strains for preventing POR [30-34]. All studies considered endoscopic as primary end point but at different time points: 3, 6, 12 or 24 months. Three studies considered also clinical as a secondary end point. None of the individual studies showed a significant effect of probiotics for preventing POR and a meta-analysis of the effects of probiotics as a class [24] suggested that their effect was not different than placebo for preventing clinical, any endoscopic and severe endoscopic. However, it should be noted that, in the meta-analysis, the pooled rate of severe endoscopic in the placebo group is 15.8% lower than expected according to that reported in a meta-analysis of the placebo rates of severe endoscopic (50.2%; 95% CI, 28% 73%) [12]. This can lead to statistical under powering. BUDESONIDE Oral budesonide has been compared to placebo for preventing POR in two RCTs. In the first study [35] 130 patients were randomised to receive controlled ileal release budesonide 6 mg daily or placebo for 12 months following ileocolonic resection: the main outcome was the proportion of patients with endoscopic (Rurgeerts score > 2) and clinical at 1 year. In the second study [36] 83 patients were randomised to receive ph dependent budesonide 3 mg or placebo for 1 year after surgical resection. The main outcome was endoscopic and clinical at 1 year. In both studies the withdrawal rate was high. Taken together (overall 212 patients, 106 receiving budesonide and 106 receiving placebo) neither the endoscopic rate (odds ratio: 0.87; 95% CI: ) nor the clinical rate (odds ratio: 0.88; 95% CI: ) was significantly reduced in budesonide-treated patients [37]. THIOPURINES Five studies evaluated the use of purine antimetabolites azathioprine (AZA) or 6-mercaptopurine (6-MP) versus mesalazine and/or placebo for preventing POR [38-42]. The studies have different design. Hanauer et al [38] randomised 131 CD patients after surgical resection, to placebo, mesalazine 3 g/daily, or 6-MP 50 mg/daily for 2 years. On intention-to-treat analysis, the 24- month clinical rates were 50%, 58% and 77% in patients receiving 6-MP, mesalazine and placebo, respectively. The corresponding figures for severe endoscopic were 16%, 48% and 42%, respectively. There are, however, 95% confidence intervals overlap for all 3 groups of treatment. Survival analyses showed lower rates of clinical relapse and overall endoscopic in patients receiving 6-MP compared to placebo but no difference was shown regard severe endoscopic. This study has several methodological problems: 1) 31% of patients withdrew from the trial for reasons other than clinical ; 2) the dose of 6-MP used (50 mg/daily) was lower than that shown to be effective for remission maintenance; 3) clinical

5 Trials in Prevention of Post Surgical Recurrence in Crohn s Disease Reviews on Recent Clinical Trials, 2012, Vol. 7, No. 4 5 was defined using an invalidated subjective score system without the use of any quantitative measure of disease activity. This arises important problems in the interpretation of the results: for example, the clinical relapse rate in the placebo group at 2 years (77%) was higher than the endoscopic rate (64%). In the study of Ardizzone et al [39], 142 CD patients were randomised to receive AZA (2 mg/kg/daily) or mesalazine (3g/daily) for 24 months after conservative surgery. Using the intention-to-treat analysis, there were 17% clinical relapses in the AZA group compared with 28% in the mesalazine group and 6% and 10% surgical relapses respectively (not significant). Furthermore survival analysis showed no significant difference in the time of clinical or surgical relapse between the study groups. In a post hoc analysis, AZA was significantly more effective than mesalazine for preventing clinical relapse in patients who had undergone more than one operation for CD. The main limitation of this trial is the lack of blinding. It should be noted that this study refers to a specific CD patient population undergoing a conservative surgical procedure (stricturoplasty and/or minimal bowel resection) for symptomatic intestinal stenoses or occlusion. D Haens et al [40] randomised 81 CD patients undergoing curative ileocecal resection and with at least 1 risk factor for (young age, active smoking, corticosteroid use in the 3 months before surgery, perforating disease or repeated resections), to receive metronidazole (250 mg tid) for 3 months and AZA ( mg/day) for 12 months or metronidazole (250 mg tid) for 3 months and placebo for 12 months. On intention to treat analysis, the 3 months severe endoscopic rate was 42.5% and 56% in the metronidazole/placebo group and metronidazole/aza group respectively (p = 0.269). The corresponding figures at 12 months were 55% and 78% respectively (p=0.035). It should be noted that the rate of severe endoscopic at 1 year in the placebo group was 78%, approximately 30% higher than expected according to the meta-.analysis of Renna et al (50.2%; 95% CI, 28% 73%) (12). It is possible than this reflects the specific high risk patient population included in the study of D Haens et al (patients with at leat 1 risk factor of POR including young age, active smoking; corticosteroid use in the 3 months before surgery, surgery for the 2nd, 3rd, or 4th resection; and perforating disease). Finally two small studies (one published only as a letter) did not provide evidence for the superiority of AZA (at a fixed dose of 50 mg/daily or at the standard dose of mg/kg/daily) over 5-ASA (3g/daily or 4 g/daily) for preventing POR [41,42]. Because of the heterogeneity of these studies, it is difficult to draw definitive conclusions regard the efficacy of thiopurines in the post-operative setting. A Cochrane metaanalysis [24] showed that AZA or 6/MP are better than placebo for preventing clinical and severe endoscopic at 12 months (Relative Risk 0.59; 95% CI 0.38 to 0.92, NNT = 7 and 0.64; 95% CI 0.44 to 0.92, NNT = 4 respectively) but no difference could be detected between thiopurines and 5-ASA (Relative Risk 1.43, 95% CI 0.95 to 2.16 for clinical and 1.54, 95% CI 0.63 to 3.79 for severe endoscopic ). A second meta-analysis was published in 2009 and included 4 RCTs (433 patients) comparing AZA or 6-MP with placebo or 5-ASA [43]. In the overall analysis, purine analogs were more effective than control arm (placebo or 5-ASA) for preventing clinical and severe endoscopic at 1 year (NNT = 13 and 7 respectively), and clinical at 2 years (NNT = 8). When performing this meta-analysis, the authors assumed that the efficacy of mesalazine for preventing POR was similar to that of placebo and, therefore, control arm included both placebo and 5-ASA groups. This assumption is not evidence-based; although the benefit is small, there is evidence that mesalazine is better than placebo for preventing POR. Although AZA and 6-MP are usually recommended for preventing POR in selected, high-risk patients, the available data do not support this position. The ECCO statement 8F Thiopurines are more effective than mesalazine or imidazole antibiotics alone for preventing both clinical and endoscopic (EL1, RGA) is not evidence-based [44] and further well-designed, placebo controlled trials are needed to define the role of thiopurines in the post-operative setting. BIOLOGIC THERAPIES A small RCT evaluated the efficacy and safety of recombinant human interleukin 10 (IL-10, Tenovil) for preventing early (12 weeks) endoscopic in CD patients who underwent curative ileal or ileocolonic resection and primary anastomosis [45]. The rationale of this study was based on the potential role for IL-10 in the prevention of inflammation and the observation that low ileal IL-10 concentrations are predictive of endoscopic [46]. Sixty-five patients were randomised to receive Tenovil 4 µg/kg once daily, Tenovil 8 µg/kg twice weekly or placebo. After 12 weeks no difference between the three groups was observed with respect to overall endoscopic and severe endoscopic. Recently a small study evaluated the role of scheduled infliximab in reducing 12 months severe endoscopic [47]. Twenty-four patients with CD who had undergone ileo-colonic resection received intravenous infliximab (5 mg/kg at 0, 2 and 6 weeks and every 8 weeks thereafter) (11 patients) or placebo (13 patients) for 12 months. The primary end point was the proportion of patients with severe endoscopic (Rutgeerts score > 2) at 1 year. The rate of endoscopic was significantly lower in the infliximab group (1 of 11 patients; 9.1%) compared with the placebo group (11 of 13 patients; 84.6%) (P =0.0006). This means a 75% absolute risk reduction of severe endoscopic and a NNT of 1.3. Although these results are impressive, the small sample size of the study weakens the conclusion. Moreover the relevant result may be amplified by the higher than expected rate in the placebo group. In fact the 1 year rate of 84.6% in the placebo group is much higher than the pooled rate of 51.5% (95% CI 46%-54%) reported in the meta-analysis of Renna et al [12]. The role of infliximab for preventing POR should therefore be confirmed in prospective RCTs with adequate sample size and definition of the patients at risk [48].

6 6 Reviews on Recent Clinical Trials, 2012, Vol. 7, No. 4 Papi et al. PREVENTION OF SYMPTOMATIC RECURRENCE IN PATIENTS WITH EARLY ENDOSCOPIC RE- CURRENCE The vast majority of RCTs evaluating different drugs for preventing POR have a similar design. Patients who have undergone radical resection with ileo-colonic anastomosis are included; randomization to the active drug or placebo occurs immediately or few weeks after surgery, the duration of the study is usually 12 to 24 months and the end point considered are endoscopic and/or clinical. Considering that the severity of early recurrent lesions at the anastomotic site is the best predictor of the subsequent clinical course [3] and symptomatic is the clinically relevant end point, recent studies have addressed the prevention of symptomatic in patients with early severe but asymptomatic endoscopic. This study design has the advantage of selecting a population of patients with high risk of developing symptoms and, therefore, the sample size is reduced. Mesalazine, AZA and infliximab have been evaluated in this setting. In a multicenter double-blind, double-dummy, randomised study, AZA mg/kg/daily has been compared with mesalazine 4 g/daily over 1 year [49]. Seventy-eight CD patients who had undergone resection with ileocolonic anastomosis in the preceding 6 to 24 months and with asymtomatic moderate to severe endoscopic (Rutgeerts score > 2), were randomised. The primary end point was therapeutic failure during 1 year, defined as clinical (CDAI score > 200 and an increase of 60 points) or drug discontinuation due to lack of efficacy or adverse events. Although there was no statistical difference between the two groups regard the primary end point (treatment failure 22.0% in the AZA group and 10.8% in the mesalazine group; difference 11.1%; 95% CI -5.0% to 27.3%, p=0.19), clinical occurred significantly less frequently in the AZA group compared to the mesalazine group (0% vs 10.8%, p=0.031) but discontinuation due to adverse drug reactions only occurred in AZA-treated patients (22.0% vs 0%, p=0.002). Significantly more patients in the AZA group showed endoscopic improvement (> 1 point reduction in Rutgeerts score) between baseline and month 12 (63.3% vs 34.4%, p=0.023). In a small open label study [50] 26 CD patients with early endoscopic 6 months after resective surgery received mesalazine 3 g/daily (n 10), AZA 50 mg/daily (n 8) or infliximab 5 mg/kg, every 8 weeks. During 6-month observation clinical occurred in 7 patients (70%) in the mesalazine group, 3 patients (38%) in the AZA group and in no patients in the infliximab group (P =0.01). impovememt occurred in 75% of patients in the infliximab group, 38% in the AZA group, and 0% in the mesalazine group (P = 0.006). CONCLUSIONS Preventing POR in CD is a complex problem. Several drugs and strategies have been evaluated to decrease the risk of in CD but the overall results are largely disappointing. Severe endoscopic and clinical are the main outcomes considered in the vast majority of RCTs. The rate (endoscopic and clinical) in the placebo group should be kept in mind when interpreting the results of RCTs and meta-analyses. Mesalazine is the drug more extensively evaluated: it is better than placebo for preventing both clinical and severe endoscopic but the absolute benefit is small. Nitroimidazole antibiotics (metronidazole and ornidazole) are effective for reducing both clinical and endoscopic, but their use as a long term treatment is limited by adverse effects. The role of thiopurines remains controversial. Although AZA, together with metronidazole, is better than placebo for reducing POR in high risk patients, the superiority of thiopurines versus mesalazine is not fully demonstrated. Infliximab reduces the risk of severe endoscopic by 75% but this data come from a single small RCT and further studies are needed. Budesonide, probiotics and interleukin 10 are ineffective for preventing POR. The strategy of preventing symptomatic in patients with early severe endoscopic should be further investigated. Preliminary data suggest the superiority of AZA and infliximab versus mesalazine in this setting. CONFLICT OF INTEREST The author(s) confirm that this article content has no conflicts of interest. ACKNOWLEDGEMENT Declared none. REFERENCES [1] Wolters FL, Russel MG, Stockbrugger RW. Systematic review: has disease outcome in Crohn s disease changed during the last four decades? Alimentary Pharmacology & Therapeutics 2004; 20: [2] Jess T, Riis L, Vind I, et al. Changes in Clinical Characteristics, Course, and Prognosis of Inflammatory Bowel Disease during the Last 5 Decades: A Population-Based Study from Copenhagen, Denmark. Inflamm Bowel Dis 2007; 13: [3] Rutgeerts P, Geboes K, Vantrappen G, et al. Predictability of the postoperative course of Crohn's disease. Gastroenterology 1990; 99: [4] Cottone M, Orlando A, Viscido A, et al. Review article: prevention of postsurgical relapse and in Crohn s disease. Aliment Pharmacol Ther 2003; 17 (Suppl. 2): [5] Borley NR, Mortensen NJ, Jewell DP. Preventing postoperative of Crohn's disease. Br J Surg 1997; 84: [6] Cottone M, Rosselli M, Orlando A, et al. Smoking habits and in Crohn s disease. Gastroenterology 1994; 106: [7] Yamamoto T, Keighley MR. Smoking and disease after operation for Crohn s disease. Br J Surg 2000; 87: [8] Reese G, Nanidis T, Borysiewicz C, et al. The effect of smoking after surgery for Crohn s disease: a meta-analysis of observational studies. Int J Colorectal Dis. 2008; 23: [9] Simillis C, Yamamoto T, Reese G, et al. Meta-Analysis comparing incidence of and indication for reoperation after surgery for perforating versus non perforating Crohn s Disease. Am J Gastroenterol. 2008; 103: [10] Yamamoto T. Factors affecting after surgery for Crohn s disease. World J Gastroenterol 2005; 11: [11] Lémann M. Review article: can post-operative in Crohn's disease be prevented? Aliment Pharmacol Ther. 2006;24 Suppl 3:22-8. [12] Renna S, Cammà C, Modesto I, et al. Meta-analysis of the placebo

7 Trials in Prevention of Post Surgical Recurrence in Crohn s Disease Reviews on Recent Clinical Trials, 2012, Vol. 7, No. 4 7 rates of clinical relapse and severe endoscopic in postoperative Crohn's disease. Gastroenterology. 2008;135: [13] Caprilli R, Andreoli A, Capurso L, et al. Oral mesalazine (5- aminosalicylic acid; Asacol) for the prevention of post-operative of Crohn's disease. Gruppo Italiano per lo Studio del Colon e del Retto (GISC). Aliment Pharmacol Ther. 1994;8: [14] McLeod RS, Wolff BG, Steinhart AH, et al. Prophylactic mesalamine treatment decreases postoperative of Crohn's disease. Gastroenterology. 1995;109: [15] Brignola C, Cottone M, Pera A, et al. Mesalamine in the prevention of endoscopic after intestinal resection for Crohn's disease. Italian Cooperative Study Group. Gastroenterology. 1995;108: [16] Florent C, Cortot A, Quandale P, et al. Placebo-controlled clinical trial of mesalazine in the prevention of early endoscopic s after resection for Crohn's disease. Groupe d'etudes Thà rapeutiques des Affections Inflammatoires Digestives (GE- TAID). Eur J Gastroenterol Hepatol. 1996;8: [17] Sutherland LR, Martin F, Bailey RJ, et al. A randomized, placebocontrolled, double-blind trial of mesalamine in the maintenance of remission of Crohn's disease. The Canadian Mesalamine for Remission of Crohn's Disease Study Group. Gastroenterology. 1997;112: [18] Lochs H, Mayer M, Fleig WE, et al. Prophylaxis of postoperative relapse in Crohn's disease with mesalamine: European Cooperative Crohn's Disease Study VI. Gastroenterology. 2000;118: [19] Hanauer SB, Korelitz BI, Rutgeerts P, et al. Postoperative maintenance of Crohn's disease remission with 6-mercaptopurine, mesalamine, or placebo: a 2-year trial. Gastroenterology 2004;127: [20] Caprilli R, Cottone M, Tonelli F, et al. Two mesalazine regimens in the prevention of the post-operative of Crohn's disease: a pragmatic, double-blind, randomized controlled trial. Aliment Pharmacol Ther. 2003;17: [21] Cammà C, Giunta M, Rosselli M, et al. Mesalamine in the maintenance treatment of Crohn's disease: a meta-analysis adjusted for confounding variables. Gastroenterology. 1997;113: [22] Cottone M, Cammà C. Mesalamine and relapse prevention in Crohn's disease. Gastroenterology. 2000;119:597. [23] Camma` C, Viscido A, Latella G, et al. Mesalamine in the prevention of clinical and endoscopic postoperative of Crohn s disease: a meta-analysis. Dig Liver Dis 2002; 34: A86. [24] Doherty G, Bennett G, Patil S, et al. Interventions for prevention of post-operative of Crohn's disease. Cochrane Database Syst Rev. 2009;CD [25] Gordon M, Naidoo K, Thomas AG, et al. Oral 5-aminosalicylic acid for maintenance of surgically-induced remission in Crohn's disease. Cochrane Database Syst Rev. 2011;CD [26] Ford AC, Khan KJ, Talley NJ, et al. P5-aminosalicylates prevent relapse of Crohn's disease after surgically induced remission: systematic review and meta-analysis. Am J Gastroenterol. 2011;106: [27] Papi C, Aratari A, Tornatore V, et al. Long-term prevention of post-operative in Crohn's disease cannot be affected by mesalazine. J Crohns Colitis. 2009;3: [28] Rutgeerts P, Hiele M, Geboes K, et al. Controlled trial of metronidazole treatment for prevention of Crohn's after ileal resection. Gastroenterology. 1995;108: [29] Rutgeerts P, Van Assche G, Vermeire S, et al. Ornidazole for prophylaxis of postoperative Crohn's disease : a randomized, double-blind, placebo-controlled trial. Gastroenterology. 2005;128: [30] Prantera C, Scribano ML, Falasco G, et al. Ineffectiveness of probiotics in preventing after curative resection for Crohn's disease: a randomised controlled trial with Lactobacillus GG. Gut. 2002;51: [31] Marteau P, Lémann M, Seksik P, et al. Ineffectiveness of Lactobacillus johnsonii LA1 for prophylaxis of postoperative in Crohn's disease: a randomised, double blind, placebo controlled GETAID trial. Gut. 2006;55: [32] Van Gossum A, Dewit O, Louis E, et al. Multicenter randomizedcontrolled clinical trial of probiotics (Lactobacillus johnsonii, LA1) on early endoscopic of Crohn's disease after lleo-caecal resection. Inflamm Bowel Dis. 2007;13: [33] Chermesh I, Tamir A, Reshef R, et al. Failure of Synbiotic 2000 to prevent postoperative of Crohn's disease. Dig Dis Sci. 2007;52: [34] Madsen K, Backer JL, Leddin D, et al. A randomized controlled trial of VSL#3 for the prevention of endoscopic following surgery for Crohn s disease. Gastroenterology 2008;134(4 Suppl 1):A361. [35] Hellers G, Cortot A, Jewell D, et al. Oral budesonide for prevention of postsurgical in Crohn's disease. The IOIBD Budesonide Study Group. Gastroenterology. 1999;116: [36] Ewe K, Böttger T, Buhr HJ, et al. Low-dose budesonide treatment for prevention of postoperative of Crohn's disease: a multicentre randomized placebo-controlled trial. German Budesonide Study Group. Eur J Gastroenterol Hepatol. 1999;11: [37] Papi C, Luchetti R, Gili L, et al. Budesonide in the treatment of Crohn's disease: a meta-analysis. Aliment Pharmacol Ther. 2000;14: [38] Hanauer SB, Korelitz BI, Rutgeerts P, et al. Postoperative maintenance of Crohn's disease remission with 6- mercaptopurine,mesalamine, or placebo: a 2-year trial. Gastroenterology. 2004;127: [39] Ardizzone S, Maconi G, Sampietro GM, et al. Azathioprine and mesalamine for prevention of relapse after conservative surgery for Crohn's disease. Gastroenterology. 2004;127: [40] D'Haens GR, Vermeire S, Van Assche G, et al. Therapy of metronidazole with azathioprine to prevent postoperative of Crohn's disease: a controlled randomized trial. Gastroenterology. 2008;135: [41] Nos P, Hinojosa J, Aguilera V, et al. Azathioprine and 5-ASA in the prevention of postoperative of Crohn's Disease. Gastroenterol Hepatol. 2000;23: [42] H Herfarth. Adverse events in clinical trials with azathioprine and mesalamine for prevention of postoperative ofcrohn s disease Gut 2006 (LETTER) [43] Peyrin-Biroulet L, Deltenre P, Ardizzone S, et al. Azathioprine and 6-mercaptopurine for the prevention of postoperative in Crohn's disease: a meta-analysis. Am J Gastroenterol. 2009;104: [44] Dignass A, Van Assche G, Lindsay JO, et al; European Crohn's and Colitis Organisation (ECCO). The second European evidencebased Consensus on the diagnosis and management of Crohn's disease: Current management. J Crohns Colitis. 2010;4: [45] Colombel JF, Rutgeerts P, Malchow H, et al. Interleukin 10 (Tenovil) in the prevention of postoperative of Crohn's disease. Gut. 2001; 49:42-6. [46] B Meresse, P Rutgeerts, H Malchow, et al. Low ileal interleukin 10 concentrations are predictive of endoscopic in patients with Crohn s disease Gut 2002;50: [47] Regueiro M, Schraut W, Baidoo L, et al. Infliximab prevents Crohn's disease after ileal resection. Gastroenterology. 2009;136: [48] Cottone M, Mocciaro F, Orlando A. Infliximab prevents Crohn's disease after ileal resection. Expert Rev Gastroenterol Hepatol. 2009;3: [49] Reinisch W, Angelberger S, Petritsch W, et al; International AZT-2 Study Group. Azathioprine versus mesalazine for prevention of postoperative clinical in patients with Crohn's disease with endoscopic : efficacy and safety results of a randomised, double-blind, double-dummy, multicentre trial. Gut. 2010;59: [50] Yamamoto T, Umegae S, Matsumoto K. Impact of infliximab therapy after early endoscopic following ileocolonic resection of Crohn's disease: a prospective pilot study. Inflamm Bowel Dis. 2009;15: