Long-term course of Crohn s disease in Japan: Incidence of complications, cumulative rate of initial

Transcription

1 Original Article: Alimentary Tract Long-term course of Crohn s disease in Japan: Incidence of complications, cumulative rate of initial surgery, and risk factors at diagnosis for initial surgery Yuho Sato 1), Toshiyuki Matsui 1), Yutaka Yano 1), Kozue Tsurumi 1), Yuki Okado 1), Yu Matsushima 1), Akihiro Koga 1), Haruhiko Takahashi 1), Kazeo Ninomiya 1), Yoichiro Ono 1), Noritaka Takatsu 1), Takahiro Beppu 1), Takashi Nagahama 1), Takashi Hisabe 1), Yasuhiro Takaki 1), Fumihito Hirai 1), Kenshi Yao 1), Daijiro Higashi 2), Kitaro Futami 2), Masakazu Washio 3) 1) Department of Gastroenterology, Fukuoka University Chikushi Hospital, Fukuoka, Japan 2) Department of Surgery, Fukuoka University Chikushi Hospital, Fukuoka, Japan 3) Department of Community Health and Clinical Epidemiology, St. Mary s College, Kurume, Japan Corresponding author: Toshiyuki Matsui 1

2 Department of Gastroenterology, Fukuoka University Chikushi Hospital Zokumyoin, Chikushino, Fukuoka , Japan Tel: ; Fax: Short title: Long-term course of Crohn s disease in Japan 2

3 Abstract Background and Aims: Intestinal complications of stenosis or fistula may occur during the course of Crohn s disease (CD), and surgery is performed in a fair number of patients. The risk factors for initial surgery in a Japanese hospital-based cohort of CD patients were evaluated. Methods: This study was a single-center, retrospective, cohort study. The subjects were 520 patients who underwent inpatient and outpatient treatment at our hospital, had a definitive diagnosis of CD, and no previous surgery. Three parameters were investigated: 1) cumulative incidence of stenosis and fistula; 2) cumulative rate of initial surgery for each disease type; and 3) risk factors at diagnosis for initial surgery. Results: 1) Stenosis and fistula increased with time, with stenosis or fistula appearing in about half of the patients after 5 years. 2) The cumulative rate of initial surgery was about 50% after 10 years. 3) The patient factors at diagnosis of current smoker, upper gastrointestinal disease, stricturing, penetrating, moderate to severe stenosis of the jejunum, moderate to severe stenosis of the ileum, and moderate to severe stenosis of the terminal ileum were risk factors for initial surgery. 3

4 Conclusions: Stenosis or fistula appeared in about half of the patients after 5 years from diagnosis. When upper gastrointestinal disease or complicated small intestinal lesions are seen at the time of diagnosis, the cumulative rate of initial surgery is significantly higher. Keywords Crohn s disease, stenosis, fistula, cumulative rate of initial surgery, risk factor for initial surgery 4

5 Introduction Crohn s disease (CD) is a chronic inflammatory disease of the gastrointestinal tract characterized by granulomatous inflammation of all layers and a discontinuous distribution. It is also known to affect all parts of the gastrointestinal tract from the mouth to the anus. The occurrence of stenosis or fistula as intestinal complications is not uncommon. These complications are especially common in long-term cases, and it is not unusual for patients to undergo frequent surgery. Since the report of Crohn et al. in 1932 [1], there have been many reports in Western countries of the long-term course associated with these intestinal complications and the rate of surgery [2-6]; whereas in Japan, this research topic has only been investigated relatively recently, and few reports have described and statistically analyzed the long-term prognosis based on course observation in a large number of patients. In this study, a database was created from long-term observations of 520 patients, and the cumulative rate of initial surgery and patient factors affecting that rate were studied to investigate the prognosis. The cumulative incidences of stenosis and fistula, the typical intestinal complications that are the main reasons for these 5

6 surgeries, were also investigated during the long-term course. 6

7 Methods Study design and patient population This study was a single-center, retrospective, cohort study. A database of all CD patients treated at the Department of Gastroenterology, Fukuoka University Chikushi Hospital was created and used in this study. All inpatient and outpatient records from 1985 to 2010 were reviewed, and the database was created by retrospectively inputting each patients parameters (approximately 100), based on level of severity. The parameters included sex, age, date of diagnosis, date of initial surgery, symptoms, Crohn s disease activity index, extra-intestinal complications, malignant tumors, imaging findings including endoscopy and X-ray examinations, severity of stenosis, fistula details, abscess, and social activity. Creation of this database took three years, from 2008 to As of May 2010, 719 patients had been registered; 167 patients who had already undergone surgery at another hospital prior to their first visit to our hospital were excluded. Moreover, since disease classifications differ in Japan, and there is no disease category of aphthae only in the Montreal classification [7], 29 patients corresponding to this category [8] were also excluded. In addition, 3 patients who 7

8 were observed for less than 3 months were excluded. The remaining 520 patients were the subjects of this study (Fig. 1). All of these patients fulfilled the revised diagnostic criteria for Crohn s disease of the Ministry of Health, Labour and Welfare [9-11] during the course of their disease, and they were definitively diagnosed. Cases that were definitively diagnosed as another disease, such as ulcerative colitis, and cases of uncertain diagnosis that did not meet the diagnostic criteria were excluded from this study. Definitions and endpoints 1) The disease behavior at the time of diagnosis was classified according to the Montreal classification [7] as inflammatory (B1), stricturing (B2), or penetrating (B3), and the cumulative incidence was calculated for each disease behavior, with the date on which the stenosis and fistula occurred or the date of final observation as the endpoint, as in the report by Cosnes et al. [12]. 2) The disease type at the time of diagnosis was classified according to the Montreal classification [7] as ileal type (L1), colonic type (L2), or ileocolonic type (L3), and the cumulative rate of initial surgery was 8

9 calculated for each disease type, with the date of initial surgery or the date of final observation as the endpoint. Surgery was defined as bowel resection or surgery for intestinal complications and did not include small operations for anal lesions or intraperitoneal drainage. Stenosis, fistula, and bleeding were the reasons for surgery. In addition, fistula included both external and internal fistulas. 3) Patient factors at diagnosis were divided into clinical characteristics, age, disease type, disease behavior, morphological changes, and initial treatment. Using these patient factors at diagnosis, the rate of initial surgery was analyzed. The following items were selected as patient factors at diagnosis. Clinical characteristics included male sex and life history, including smoking and drinking; and diagnosis involved, whether or not it was made at our hospital, which specializes in CD. Smokers were defined as current smokers. Alcohol drinkers were defined as those who drank alcohol almost every day. Age, disease type, and disease behavior were classified according to the Montreal classification [7]. Upper gastrointestinal disease (L4) with CD was defined using the Japanese criteria [7-9]: 1. longitudinal ulcer or cobblestone appearance; 2. irregular to 9

10 oval-shaped ulcers or aphthae over a wide area; 3. erosions and aphthae with a trend to a longitudinal array; and 4. bamboo joint-like appearance and notch-like depressions. The lesions fulfilled at least one of the criteria listed above (1-4) in the esophagus, stomach, or duodenum. However, the above applied only to patients in whom other diseases could be excluded. Morphological changes included segments in which moderate to severe stenosis had occurred, with the small intestine divided into jejunum, ileum, and terminal ileum; and segments in which moderate to severe stenosis had occurred, with the colon divided into the three segments of the left colon, right colon, and rectum. Moderate to severe stenosis was determined by double-contrast imaging with a sufficient amount of injected air, and stenosis was defined as stenosis in which the lumen was less than one-half that of neighboring healthy intestine. The initial treatment was the treatment that was provided at the time of diagnosis, and included cases of treatment with more than one agent. Statistical analysis The data were analyzed using Stat View (Stat View for Windows, version 5.0, SAS Institute Inc.). The 10

11 Kaplan-Meier method was used to compare the cumulative incidences of stenosis and fistula and the cumulative rate of initial surgery in each time period, and patient factors affecting initial surgery were analyzed using the Cox proportional hazards model. The log-rank test (Mantel-Cox) was used in tests of significance, with a probability level of 5% indicating a significant difference in all cases. 11

12 Results Table 1 shows the clinical features of the 520 patients with CD. The sex ratio (male: female) was 2.4:1, the mean age at diagnosis was 25.2 ± 10.0 years, and the mean duration of disease was 10.3 ± 7.8 years. The mean observation period (time from diagnosis until first surgery or until final observation date) was 6.6 ± 6.0 years. Overall, 200 patients were current smokers, 165 were alcohol drinkers. Seventy-seven patients had an anal fistula; and L1, L2, L3, and L4 were seen in 199, 84, 237, and 227 patients, respectively. The disease behavior was B1 in 422 patients, B2 in 72 patients, and B3 in 26 patients. Including some cases of multiple treatment, the treatment at the time of diagnosis was 5-aminosalicylate in 251 patients, elemental diet (> 900 kcal/day) in 395 patients, oral corticosteroid in 84 patients, 6-mercaptopurine or azathioprine in 30 patients, and biological agents in 12 patients. 1) Cumulative incidence of stenosis and fistula The cumulative incidence for each disease behavior is shown in Fig. 2. The incidence for each disease behavior at baseline from the day of diagnosis was 81% for B1, 14% for B2, and 5% for B3. The cases were 12

13 mostly B1 at baseline, but after 5 years, the percentages changed to 46%, 35%, and 19%, respectively, for B1, B2, and B3; with B2 and B3 increasing thereafter. 2) Cumulative rate of initial surgery for each disease type There were 277 surgery patients, in whom stenosis, fistula, and bleeding were the main reasons for initial surgery (Table 2). Figure 3 shows the cumulative rate of initial surgery at 5, 10, 15, and 20 years from the time of diagnosis for all 520 patients and patients of each of the three disease types. In all 520 patients, the rate was 32% after 5 years, 55% after 10 years, 70% after 15 years, and 82% after 20 years. By disease type, the rate for L1 was 37% after 5 years, 60% after 10 years, 73% after 15 years, and 85% after 20 years. For L2, it was 19%, 37%, 43%, and 48%, respectively, and for L3, it was 31%, 55%, 73%, and 86%, respectively. In a comparison of L1 and L2, the rate was significantly (p < 0.01) lower with L2. Even between L3 and L2, the rate was significantly (p < 0.05) lower with L2. In contrast, there was no significant difference between L1 and L3. 3) Risk factors at diagnosis for initial surgery Table 3 summarizes the relative risk of initial surgery according to patient factors at diagnosis on univariate 13

14 analysis. Among the patients clinical characteristics, male patients showed an increased age-adjusted risk of initial surgery than female patients, but male sex failed to be a significant risk factor for initial surgery after additional adjustment for smoking and drinking. On the other hand, current smokers showed a significantly higher risk than former smokers and non-smokers even after adjusting for age, sex, and drinking. For age at the diagnosis of CD, there was no meaningful association between age and the risk of initial surgery. Among patient factors of disease type, patients with L2 showed a significantly lower age, sex, smoking, and drinking-adjusted risk of initial surgery than patients with L1, while compared with their counterparts, those with L4 showed a significantly higher risk of initial surgery even after controlling for age, sex, smoking, and drinking. Among patient factors of disease behavior, patients with B2, as well as those with B3, showed a significantly increased age, sex, smoking, and drinking-adjusted risk of initial surgery than those with B1. Among patient factors of morphological changes (i.e., moderate to severe stenosis), those with jejunal stenosis showed a significantly higher risk of initial surgery than those without, even after adjusting for age, sex, smoking, and drinking. Similar findings were observed among those with ileal stenosis and those with terminal 14

15 ileal stenosis. Among patient factors of initial treatment, biological agents showed a hazard ratio higher than one, while oral corticosteroid and 6-mercaptopurine or azathioprine showed a hazard ratio lower than one, but none of them were significant. The relationship with L4 is shown in Table 4, and that with other factors for which there was a significant difference is shown in Table 3. The results show that patients with L4 at the time of diagnosis had a significantly higher percentage of B2 and moderate to severe stenosis of the jejunum at diagnosis. 15

16 Discussion The long-term course of intestinal complications and the surgery rate for CD has been presented in many reports [1-6] in Western countries. However, few reports have described the long-term course in a large number of patients in Asia or in Japan [13, 14]. Therefore, this was investigated in this study. The present results suggested that intestinal complications of stenosis and fistula were the main reasons for intestinal surgery in CD. Stenosis and fistula increased with time, with stenosis or fistula appearing in about half of the patients after 5 years. As may be conjectured from the increase in intestinal complications with time, the cumulative rate of initial surgery also increased with time. The majority of patients had undergone surgery at 10 years after diagnosis. This result is similar to reports from Western countries [2, 6, 15, 16], and the prognosis of CD in Japan was no better than elsewhere. In addition, there was no considerable difference in the cumulative rate of initial surgery even over the course from onset [13, 14]. The cumulative rate of initial surgery was also investigated for each disease type. As shown in Fig. 3, the 16

17 cumulative rate of initial surgery was significantly lower for L2 than for the other disease types. This may also be due to a tendency for surgery to be more likely in cases with small bowel lesions, which was also pointed out in a report by Sands et al. [12, 17] showing that ileal type at the time of diagnosis is a risk factor for surgery. One may understand that the lumen of the small bowel is smaller than that of the large bowel, and the formation of a fistula from stenosis or intestinal obstruction from the stenosis itself is more likely to occur. Regarding the risk for L2, there are several reports of a lower cumulative rate of surgery with L2 than with other disease types in Western countries [15, 18-20]. How patient factors at diagnosis affect this initial surgery was also investigated in this study. The analysis was done for 25 items of clinical characteristics, age, disease type, disease behavior, morphological changes, and initial treatment. Seven factors at diagnosis (current smoker, L4, B2, B3, moderate to severe stenosis of the jejunum, moderate to severe stenosis of the ileum, moderate to severe stenosis of the terminal ileum) were identified as independently associated with an increased risk of initial surgery. Smoking has also been identified as a significant risk factor, which agrees with previous reports [14, 16, 17]. 17

18 Regarding morphological features, upper gastrointestinal disease involvement was identified, but upper gastrointestinal disease involvement has reported as a risk in few investigations to date. Peyrin-Biroulet et al. [16] recently reported upper gastrointestinal disease as a risk factor in addition to factors such as male sex, smoking, and anal fistula for the cumulative incidence of initial surgery from diagnosis. High frequencies of upper gastrointestinal lesions (52-92%) have been reported both in Japan and other countries [21-25], and a similar frequency of upper gastrointestinal lesions was seen in the present study. Moreover, in the proposed diagnostic criteria for CD in Japan [9-11], characteristic gastric and duodenal lesions are given as a secondary diagnostic finding in addition to main findings, and they have come to occupy a more important position even in the diagnosis of upper gastrointestinal disease. As shown in Table 4, patients with L4 at the time of diagnosis included many who also had B2 and moderate to severe stenosis of the jejunum among patient factors at diagnosis. From this result, it may be presumed that patients with L4 at the time of diagnosis are a high risk group for initial surgery. While Beaugerie et al. [26] and Peyrin-Biroulet et al. [16] reported that anal fistulas were a factor related to 18

19 poor prognosis, in the present study, anal fistula was not a risk factor for initial surgery. The reason for this may be that patients visited our hospital with symptoms such as abdominal pain, fever, or anal discomfort before the occurrence of anal fistula. In other words, it may be presumed that the percentage of cases in which anal fistula could be diagnosed at the time of diagnosis was low because patients presented or were referred to our hospital with milder anal lesions. Therefore, while anal fistula may be a risk factor for surgery over the long term, anal fistula at the time of diagnosis does not seem to be a risk factor for initial surgery. The present study has several limitations. First, this study had a long-term course including various treatments, and therapeutic methods over the long-term period were not considered. In particular, there are a number of reports showing improvement in the rate of surgery with biological agents in recent years. However, biological agents were used as maintenance therapy from the time of initial diagnosis in only 12 patients, and no significant differences were seen. Second, there were no detailed data on smoking and drinking. In conclusion, stenosis or fistula appeared in about half of the patients after 5 years. When upper 19

20 gastrointestinal disease or small intestinal lesions are seen at the time of diagnosis, the cumulative rate of initial surgery is significantly higher. Acknowledgments This work was supported by Health and Labour Sciences Research Grants for Research on Intractable Disease from the Ministry of Health, Labor and Welfare of Japan. Conflict of interest The authors declare that they have no conflict of interest. 20

21 References 1. Crohn BB, Ginzburg D, Oppenheimer GD, Regional ileitis: A pathologic and clinical entity. JAMA 1932, 99: Peyrin-Biroulet L, Loftus EV Jr, Colombel JF, Sandborn WJ. The natural history of adult Crohn s disease in population based cohorts. Am J Gastroenterol 2010; 105: Lakatos PL, Golovics PA, David G, Pandur T, Erdelyi Z, Horvath A, et al. Has there been a change in the natural history of Crohn s disease? Surgical rates and medical management in a population based inception cohort from Western Hungary between Am J Gastroenterol 2012; 107: Picco MF, Zubiaurre I, Adluni M, Cangemi JR, Shelton D. Immunomodulators are associated with a lower risk of first surgery among patient with non-penetrating non-stricturing Crohn s disease. Am J Gastroenterol 2009; 104: Thia KT, Sandborn WJ, Harmsen WS, Zinsmeister AR, Loftus EV Jr. Risk factors associated with progression to intestinal complications of Crohn s disease in a population based cohort. Gastroenterology 21

22 2010; 139: Bernell O, Lapidus A, Hellers G. Risk factors for surgery and postoperative recurrence in Crohn s disease. Ann Surg 2000; 231: Silverberg MS, Satsangi J, Ahmad T, Arnott ID, Bernstein CN, Brant SR, et al. Toward an integrated clinical, molecular and serological classification of inflammatory bowel disease: Report of a Working Party of the 2005 Montreal World Congress of Gastroenterology. Can J Gastroenterol 2005; 19 (Suppl. A): Tsurumi K, Matsui T, Hirai F, Takatsu N, Yano Yutaka, Hisabe T, et al. Incidence, clinical characteristics, long-term course, and comparison of progressive and non-progressive cases of aphthous-type Crohn s disease: a single-center cohort study. Digestion 2013; 87: Matsui T, Hirai F, Hisabe T. Proposed diagnostic criteria for Crohn s disease. Annual of Research Group of Intractable Inflammatory Bowel Disease subsidized by the Ministry of Health, Labour, and Welfare of Japan. 2011: (in Japanese). 22

23 10. Ueno F, Matsui T, Matsumoto T, Matsuoka K, Watanabe M, Hibi T. Evidence-based clinical practice guidelines for Crohn s disease, integrated with formal consensus of experts in Japan. J Gastroenterol 2013; 48: Hisabe T, Matsui T, Hirai F, Watanabe M. Evalution of diagnostic criteria for Crohn s disease in Japan. J Gastroenterol 2014; 49: Cosnes J, Cattan S, Blain A, Beaugerie L, Carbonnel F, Parc R, et al. Long-term evolution of disease behavior of Crohn s disease. Inflamm Bowel Dis 2002; 8: Gao X, Yang RP, Chen MH, Xiao YL, He Y, Chen BL, et al. Risk factors for surgery and postoperative recurrence: analysis of a south China cohort with Crohn's disease. Scand J Gastroenterol 2012; 47: Song XM, Gao X, Li MZ, Chen ZH, Chen SC, Hu PJ, et al. Clinical features and risk factors for primary surgery in 205 patients with Crohn's disease: analysis of a South China cohort. Dis Colon Rectum 2011; 54:

24 15. Lind E, Fausa O, Gjone E, Mogensen SB. Crohn s disease. Treatment and outcome. Scand J Gastroenterol 1985; 20: Peyrin-Biroulet L, Harmsen WS, Tremaine WJ, Zinsmeister AR, Sandborn WJ, Loftus EV Jr. Surgery in a population-based cohort of Crohn's disease from Olmsted county, Minnesota ( ). Am J Gastroenterol 2012; 107: Sands BE, Arsenault JE, Rosen MJ, Alsahli M, Bailen L, Banks P, et al. Risk of early surgery for Crohn's disease: implications for early treatment strategies. Am J Gastroenterol 2003; 98: Farmer RG, Whelan G, Fazio VW. Long-term follow-up of patients with Crohn s disease. Gastroenterology 1985; 88: Harper PH, Fazio VW, Lavery IC, Jagelman DG, Weakley FL, Farmer RG, et al. The long term outcome in Crohn s disease. Dis Colon Rectum 1987; 30: Mekhjian HS, Switz DM, Watts HD, Deren JJ, Katon RM, Beman FM. National cooperative Crohn s disease after surgery. Gastroenterology 1979; 77:

25 21. Matsumura M, Matsui T, Hatakeyama A, Matake H, Uno H, Sakurai T, et al. Prevalence of Helicobactor pylori infection and correlation between severity of upper gastrointestinal lesions and H. pylori infection in Japanese patients with Crohn s disease. J Gastroenterol 2001; 36: Cameron DJ. Upper and lower gastrointestinal endoscopy in children and adolescents with Crohn s disease: a prospective study. J Gastroenterol hepatol 1991; 6: Nugent FW, Roy MA. Duodenal Crohn s disease: an analysis of 89 cases. Am J Gastroenterol 1989; 84: Kundhal PS, Stormon MO, Zachos M, Critch JN, Cutz E, Griffiths AM. Gastral antral biopsy in the differentiation of pediatric colitides. Am J Gastroenterol 2003; 98: Castellaneta SP, Afzal NA, Greenberg M, Deere H, Davies S, Murch SH, et al. Diagnostic role of upper gastrointestinal endoscopy in pediatric inflammatory bowel disease. J Pediatr Gastroenterol Nutr 2004; 39: Beaugerie L, Seksik P, Nion-Larmurier I, Gendre JP, Cosnes J. Predictors of Crohn s disease. 25

26 Gastroenterology 2006; 130:

27 Figure legends Figure 1. Of 719 hospitalized patients as of May 2010, 520 patients remained after exclusion criteria were applied. One hundred sixty-seven patients who had already undergone surgery at another hospital prior to their first visit to our hospital and 29 patients with aphthae only were excluded. Another 3 patients who were observed for less than 3 months were excluded. Figure 2. Cumulative incidences of fistula (upper solid line) and stenosis (lower broken line) from the time of diagnosis in 520 patients. The follow-up number over time is also shown. Figure 3. Cumulative rate of initial surgery for each disease type. Ileal type (L1) and ileocolonic type (L3) have significantly higher rates than colonic type (L2). 27

28 28

29 29

30 30

31 31

32 32

33 33

34 34