Risk Factors For Primary Surgery In Patients With Crohn’s Disease:a Cohort Study Of315Cases

Background and purposeCrohn’s disease (CD) is one of the major subtype of inflammatory bowel disease (IBD), characterized by a chronic relapsing, inflammatory granulomatous and segmental distribution of lesions. It could affect the whole gastrointestinal tract, with the distal ileum and colon in neighbouring most common seen. Crohn’s disease is quite common in Europe and the United States, and its annual incidence is approximately6-15per million, and prevalence50-200per million. A newly published epidemiological study conducted by multi-center collaborations in the Asia-Pacific region showed that during time from April1,2011to March31,2012, the average annual incidence of Crohn’s disease in Asian region is0.54per million. Although the frequency of Crohn’s disease is still quite loewer in Asia than in the West, the epidemiological studies in recent20years showed that the prevalence Crohn’s disease in Asia is on the rise.The cause of Crohn’s disease is complex, and the pathogenesis is still unclear. View in recent years holds that genetic, environmental factors and immune factors are related to the pathogenisis of Crohn’s disease. The host carrying susceptibile genes, with the participation of environmental factors and immune dysfunction, eventually leads to the pathogenisis of the disease. Population-based epidemiology in western countries proved that the positive rate of family history of Crohn’s disease as high as13.4%-18.0%among patients with Crohn’s disease, and first-degree relatives of patients with Crohn’s disease present6-10times higher risk than the general population. The positive rate of family history of Crohn’s disease reported in Japan, Korea, Hong Kong and China mainland is about1.51-2.80%, significantly lower than that in the West. At the same time, the molecular genetics studies in Europe and the US have proved that the NOD2/CARD15gene on the16th chromosome is highly correlated to Crohn’s disease among the white, yet this phenomenon is not found in the Asian population, which indicates that there may exist difference on susceptible genes of Crohn’s disease between the Asian populations and the white.In addition to the difference in genetic background, the clinical features and the natural course of Crohn’s disease in Asian populations may also vary from that in the West. Presented at the World Congress of Gastroenterology in2005, Montreal classification of Crohn’s disease provided us a very reliable standard for comparing the features of Crohn’s disease between different groups. At the same time, the surgery rate in patients with Crohn’s diease is very high throughout the course of the disease. Studies have shown that up to70-90%of the patients suffering from Crohn’s disease will eventually require surgical treatment because of the occurrence of intestinal obstruction, perforation, fistula, abdominal abscess, gastrointestinal bleeding and other serious complications. The procedure of surgery is simply to control symptoms and recover the function of bowel, but not a cure to the disease. Meanwhile, about half of the patients required reoperation. The need of repeated surgery seriously harms the patients’physical and mental health.Currently the studies related to the risk factors for surgery in patients with Crohn’s disease are rare. Several researches showed that NOD2/CARD15-positive, age at diagnosis, structuring, penetrating, upper gastrointestinal tract involvement, extraintestinal manifestations, perianal lesions, and smoking habit were considered to be the risk factors for surgery in patients with Crohn’s disease, while the results were not consistent in these studies. At present, studies on Crohn’s disease in China are mostly clinical case reports, researches on the clinical characteristics, Montreal classification and risk factors for surgery is relatively rare. Meanwhile, it is of great importance for clinical practice to verify the risk factors for surgery in patients with Crohn’s disease. Studying the pattern of natural course of Crohn’s disease and determining the risk factors for surgery in patients with Crohn’s disease can help the clinicians to filter the patients at high risk for surgery, and develop a personalized treatment plan for the patient exposed to risk factors, which may reduce the operation rate in patients with Crohn’s disease, resulting greater benefits gained from the treatment.Based on the above considerations, we designed a retrospective historical cohort study in a single Center in South of China, aimed at understanding the clinical characteristics, Montreal classification and cumulative surgery rate in patients with Crohn’s disease, exploring the potential risk factors for surgery in patients with Crohn’s disease, providing an objective basis for clinicians to develop more beneficial schemes to patients suffering from Crohn’s disease, as well as providing reference data for the prospective studies with multi-center collaborations in the future.MethodMedical notes of consecutive patients with Crohn’s disease but without bowel resection history, hospitalized in Nanfang hospital, the first affiliacted hospital of Southern Medical University from January1,2003to December31,2012were retrospectively reviewed. Data of general demographic information, clinical manifestations, laboratory tests, imaging, endoscopy, pathological findings, medications, surgery, and the telephone follow-up of patients who did not perform surgery during hospitalization was carefully recorded.The onset of symptoms was considered to be the entry point. The date of primary bowel resection or the last follow-up time for the patients who did not perform bowel surgery during the whole research time was defined the end point of the study. The course of the disease is defined as the interval between the entry point and the end point. Kaplan-Meier method was applied to estimate the cumulative operation rate and log-rank test was used to test the homogeneity of various clinical variables. Multivariate Cox proportional hazards regression analysis was performed to identify independent risk factors for primary bowel resection. Effects of potential risk factors were quantified by calculating the hazard ratio (HR) and confidence interval (CI) from the final Cox model. A P value less than0.05was considered statistically significant.Results1. A total of315patients with Crohn’s disease were included in this study, the ratio of male to female was2.6:1(228/87),91.9%of the patients came from South China. Median diagnosis age (32y,9-69y) was4years older than the median age of onset (28y,8-63y).22patients (7.0%) had previous history of appendectomy,30cases (9.5%) were smokers and all of whom wew male,2patients (0.6%) had positive family history of Crohn’s disease. Abdominal pain (87.6%), diarrhea (65.4%), weight loss (54.0%) were the three predominant symptoms in patients with Crohn’s disease.2. According to Montreal classification of Crohn’s disease, the main type of age at diagnosis was A2(17-40years,66.0%); disease location mainly L1(ileal,50.5%), followed by L3(ileocolonic,34.9%), L2(colon,14.6%) was relatively uncommon.4.8%patients had upper gastrointestinal involvement (L4). Disease behavior was classified as non-stricturing non-penetrating (B1,56.2%), followed by stricutring(B2,36.5%) penetrating (B3,7.3%).18.1%patients had perianal disease (P).3. The median duration of the patients with Crohn’s disease included in this study was4years (1-20years). During the whole duration of the disease, a total of91patients (28.9%) underwent primary bowel resection,14.3%of the operation patients required emergency surgery. The surgical site mainly located in "the small intestine (48.4%)" and "small intestine and Colon (46.2%)", the proportion of colon resection (5.5%) is low. The causes for the primary surgery in patients with Crohn’s disease were as following:intestinal obstruction (52cases,57.14%), hamorrhage (11cases,12.09%), perforation (6cases,6.59%), fistula (6cases,6.59%), abodominal mass (6casese,6.59%), intestinal giant ulcer with no response to medicatications (6cases,6.59%) and others (4cases,4.4%)。4. Univariate analysisKaplan-Meier analysis showed that the overall cumulative rate of primary bowel surgery was12%at1year after onset of symptoms,17%at2years,26%at5years,46%at10years, and82%at20years. Kaplan-Meier method was used to the analysis of clinical categorical variables, estimate the cumulative operation rate in5years, log-rank test was applied to test the effects of various clinical categorical variables on the cumulative operation rate. Univariate analysis showed that:(1) The following clinical variables were of no statistically significant differences in relation to the effect on the cumulative operation rate in patients with Crohn’s disease (p>0.05):age at diagnosis (5-year cumulative operation rate:40.0%among patients≤16years old,30.1%among patients17-40years old,36.7%among patients≥40years old, p=0.852), upper gastrointestinal tract involvement modifier L4(5-year cumulative operation rate:32.8%among patients without involvement of upper gastrointestinal tract,28.6%among patients with Involvement of upper gastrointestinal tract, p=0.813), The history of appendectomy (5-year cumulative operation rate:33.3%among patients without history of appendectomy,25.0%among patients with a history of appendectomy, p=0.499), extraintestinal manifestations (5-year cumulative operation rate:32.5%among patients without extraintestinal manifestations,33.3%among patients with extraintestinal manifestations, p=0.740).(2) The following clinical variables are associated with elevated primary bowel operation rate:gender (5-year cumulative operation rate:34.3%for men,28.0%for women, p=0.040), disease location (5-year cumulative operation rate:37.5%for L1,12.0%for L2,32.8%for L3, p=0.015), disease behavior(5-year cumulative operation rate:12.3%among patients with non-stricturing non-penetrating disease,49.5%among patients with stricturing disease, and38.1%%among patients with penetrating disease; p<0.001), smoking (5-year cumulative operation rate:61.9%among smokers,29.1%among non-smokers; p=0.002).(3) The following clinical variables were associated with reduced primary bowel operation rate:perianal disease modifier P (5-year cumulative operation rate:10.7%among patients with perianal lesions,36.4%among patients without perianal lesions; p=0.022). In addition, The use of any kind of the following medication for the treatment of Crohn’s disease significantly reduce the primary bowel operation rate in patients with Crohn’s disease (p<0.05):5-aminosalicylic acid compounds (5-year cumulative operation rate:68.4%among patients without the use of5-aminosalicylic acid compounds,17.6%among patients with the use of5-aminosalicylic acid compounds; P<0.001), glucocorticoids (5-year cumulative operation rate:50.5%among patients without the use of GC,13.8%among patients with the use of GC; P<0.001), immunosuppressants (5-year cumulative operation rate:39.0%among patients without the use of immunosuppressants,2.9%among patients with the use of immunosuppressants; P<0.001), anti-TNF-a biologics (5-year cumulative operation rate:35.3%among patients without the use of anti-TNF-a biologics,13.0%among patients with the use of anti-TNF-a biologics; P=0.003).4. Multivariate analysisIn the final Cox Proportional hazards regression model, all the clinical variables were put into the multivariate analysis model, wald test was used to test if the various clinical variables have no statistically influence on the primary bowel operation rate. In this Cox multivariate model, stricturing(HR:4.803;95%CI:2.466-9.352; P<0.001), penetrating (HR:13.197;95%CI:5.938-29.330; P<0.001), smoking (HR:1.932;95%CI:1.031-3.619; P=0.041) were associated with an increased risk for primary bowel operation, and perianal disease modifier P (HR:0.027;95%CI:0.193-0.889; P=0.024), the prescription of5-aminosalicylic acid compounds (HR:0.277;95%CI:0.162-0.474; P<0.001) were associated with a significantly decreased primary bowel operation rate.ConclusionIn this cohort study of a single center in South China, we found that male patients had a predominance, patients with the age of17-40years old were mostly common seen in Crohn’s disease, disease location mainly for L1(The ileum), followed by L3(Indiana), the main disease behavior was B1(non-stricturing non-penetrating). In the development process, the cumulative bowel operation rate increased yearly in patients with Crohn’s disease, stricturing, penetrating behavior and smoking habit were independent risk factors for primary bowel surgery, priscription of aminosalicylic acid compounds and contaminant perianal disease were associated with reduced rate of the primary bowel surgery.