1 Introduction

Inflammatory bowel disease (IBD), including Crohn's disease (CD), ulcerative colitis (UC), and inflammatory bowel disease unclassified (IBD-U), is an idiopathic chronic inflammatory disorder, which has traditionally been regarded as common disease of the West. With the continuous development of newly industrialized countries, IBD has emerged globally with high incidence and prevalence [1, 2]. In China, it is estimated that there will be more than 1.5 million cases of IBD by 2025, which may pose a substantial and significant disease burden [2].

A comprehensive understanding of IBD clinical phenotypes will provide insight into the key factors that trigger disease onset and progression, which in turn can help identify new therapeutic targets and design preventive approaches. With the advances of artificial intelligence and big data technology, clinical databases have become an important cornerstone and tool for obtaining epidemiological and clinical clues [3].

Since IBD first appeared in Western countries, there have been more clinical data collections on IBD in these countries, including commercial databases such as the Optum Research Database [4] and the Vizient Clinical Database, as well as medical databases such as The Health Improvement Network (THIN) [5] and the Manitoba IBD Epidemiology Database [6]. These clinical databases integrated the medical information of IBD patients. Combined with big data and advanced analytics, the databases revealed patterns about disease development, treatment, and prognosis to guide clinicians and policymakers in practice.

To translate the existing IBD clinical information into medical research resources, the Asia-Pacific region, including China, has also established its own IBD clinical databases [7]. The IBD clinical databases in the Asia-Pacific region include the database of the Korean National Health Insurance Claims [8], the Clalit Health Services Database [9], and the Health Insurance Research Database in Taiwan area [10]. In China, some institutions have built their databases with less than 100 items for Chinese IBD patients. Unfortunately, these databases lack uniformity in data entry fields, making it difficult to both share and perform epidemiological studies between institutions. Moreover, there is no unified IBD database where clinical characteristics and treatment modalities in both Eastern and Western countries have been integrated, making studies between Asian and Western countries challenging.

To facilitate the effective integration of information between clinical databases, we developed items specific to Chinese IBD clinical features on the basis of the Penn IBD database with the North American IBD clinical characteristics and named the 301 IBD database. Then, we collected and organized clinical data from IBD inpatients from the First Medical Center of Chinese PLA General Hospital (PLAGH), between 2008 and 2023 based on new database entries to validate this database.

2 Materials and Methods

3 Results

3.1 Construction of the 301 IBD Database

The 301 IBD database which referred to the Penn IBD database consist of six sections, including demographic characteristics, personal history, treatment, laboratory tests and examinations, clinical phenotype, and disease activity. The 301 IBD database included 490 items, in which 241 were from the Penn IBD database and 249 newly added items based on the IBD guidelines and consensus of China (Figure 1 and Table S1).

The main additions in the new database are as follows: (1) increase the causes of disease onset and flare-up; (2) add disease-related biochemical indicators, especially opportunistic infection screening; (3) focus on UC complications, such as toxic megacolon, intestinal perforation, and so forth; (4) record hospitalization costs; (5) nonalcoholic fatty liver disease and cholelithiasis were added to the extraintestinal manifestations (EIMs), and perianal disease was added to the UC module; and (6) the treatment section adds fecal microbiota transplantation (FMT), traditional Chinese medicine, and probiotics (Table 1). Features of the 301 IBD database include inpatient focus, biochemical indicators and opportunistic infection focus, and more about UC-associated complications.

Table 1. Comparison of IBD items from the PLAGH and Penn Medicine.

Items	301 IBD database	Penn IBD database
Demographic characteristics	Diagnosis Onset and definite time Manifestation Hospitalization	Diagnosis Onset and definite time
Personal history	EIMs history Antibiotics NSAIDs	EIMs history Narcotic pain medications
Treatment	Aminosalicylic acid Glucocorticoid Immunosuppressant Biologic agent Traditional Chinese medicine Probiotics FMT	Aminosalicylic acid Glucocorticoid Immunosuppressant Biologic agent
Laboratory tests	Test for hepatitis virus Quantiferon TPMT Test for EBV Test for CMV Test for opportunistic pathogen	Test for hepatitis virus Quantiferon TPMT
Examinations	Endoscopy Abdominal CT scan Abdominal MRI scan DXA score MRI of spine and joints	Endoscopy Abdominal CT scan Abdominal MRI scan DXA score
Clinical phenotype	CD phenotype UC phenotype IBD-U phenotype Complications (colorectal cancer, toxic megacolon, intestinal obstruction, intestinal perforation, cervical dysplasia, ovarian cancer, and skin cancer)	CD phenotype UC phenotype IBD-U phenotype Complications (colorectal cancer, cervical dysplasia, and skin cancer)
Disease activity	Defecation situation Abdominal symptoms Extraintestinal manifestation Psychological situation	Defecation situation Abdominal symptoms Extraintestinal manifestation Psychological situation

• Abbreviations: CMV, cytomegalovirus; DXA, dual energy x-ray absorptiometry; EBV, Epstein–Barr virus; EIM, extraintestinal manifestation; FMT, fecal microbiota transplantation; NSAIDs, nonsteroidal anti-inflammatory drugs; TPMT, thiopurine methyltransferase.

3.2 Single-Center UC Clinical Data Analysis

3.2.1 Demographic Characteristics

A total of 1363 cases of IBD, with 1053 UC cases (1944 hospitalizations) and 310 CD cases (662 hospitalizations) were admitted to the PLAGH from January 1, 2008 to 31 December 31, 2023. During the period, composition ratio of UC patients in the gastroenterology ward rose significantly, from 1.12% in 2008 to 2.25% in 2023 (Z = 30.673, p< 0.001), whereas, CD held relatively steady, from 0.51% in 2008 to 1.33% in 2023 (Z = 0.917, p= 0.338) (Figure 2A). The composition ratio of total IBD patients in the gastroenterology ward showed an increasing trend during the 16 years, from 2.35% to 3.94%.

For UC, the male sex was predominant, and the sex ratio was 1.67:1 (658/395). The age-specific distribution analysis showed a unimodal pattern of disease onset age, with a slight difference between the sexes. The peak age range of onset for males was 21–30 years, while for females, it was 31–40 years (Figure 2B). The average age of onset was 40.02 ± 15.44 years and diagnosis was 40.99 ± 15.43 years. The mean disease duration was 5.19 ± 6.96 years. There was no difference in onset age, diagnosis age, and disease duration between the genders (p = 0.772, 0.795, and 0.056).

For CD, the sex ratio was 2.16:1 (212/98). The peak age of onset for males was 21–30 years and for females was 41–50 years (Figure 2C). The average age of onset was 35.43 ± 14.75 years and diagnosis was 38.22 ± 15.90 years. The mean disease duration was 4.11 ± 4.29 years. The age of onset and diagnosis in males was lower than that in females, and the difference was statistically significant (p = 0.001, < 0.001), but there was no statistically significant difference in disease course (p = 0.353).

3.2.2 Clinical Phenotypes of UC Patients

The lesion extent of UC, according to the Montreal classification [17], was reported in 930 patients. More than half (55.05%) of the UC hospitalized patients had extensive colitis. About 28.60% had left-sided colonic involvement, and 14.73% had proctitis. In addition to the three phenotypes of UC, we observed that 1.61% of patients had only right-sided colonic involvement. Of the 1053 patients, 944 were in the active stage, and 109 were in remission. Among the active group, 453 cases were mild, 191 were moderate, and 300 were severe. Mucosal remission was observed in 32 of the 109 patients in the remission stage. Table 2 shows the disease phenotypes of all patients.

Table 2. Disease phenotypes of patients with UC.

Phenotypes	Cases (n)	Percentage (%)
Clinical classification
Initial type	253	24.03
Chronic recurrent type	800	75.97
Clinical disease severity
Remission	109	10.35
Mild active	453	43.02
Moderate active	191	18.14
Severe active	300	28.49
Endoscopic lesion extent
E1 (proctitis)	137	14.73
E2 (left-sided colitis)	266	28.60
E3 (extensive colitis)	512	55.05
Right-sided colitis	15	1.61

• Abbreviation: UC, ulcerative colitis.

EIMs were found in 164 patients (15.57%), and 41 patients (3.89%) presented two or more EIMs. No gender differences in the prevalence of EIMs were observed (p = 0.579). Arthritis was the most common EIM, which accounted for 7.69% of patients.

3.2.3 Clinical Phenotypes of CD Patients

Among the patients with CD, the proportion of L1, L2, L3, and L4 was 36.45%, 22.58%, 33.55%, and 7.42%, respectively. The most common disease behavior was B1, with 66.45% of patients. Concomitant perianal disease was present in 17.74% of patients (Table 3).

Table 3. Disease phenotypes of patients with CD.

Phenotypes	Cases (n)	Percentage (%)
Location
L1 (ileal)	113	36.45
L2 (colonic)	70	22.58
L3 (ileocolonic)	104	33.55
L4 (isolated upper disease)	23	7.42
Behavior
B1 (nonstructuring, nonpenetrating)	206	66.45
B2 (structuring)	82	26.45
B3 (penetrating)	22	7.10
Perianal disease	55	17.74

• Abbreviation: CD, Crohn's disease.

EIMs were found in 86 patients (27.74%), which is more common than UC (p < 0.01). Aphthous ulcer was the most common EIM, which accounted for 13.23% of patients (Table 4).

Table 4. Extraintestinal manifestations of IBD.

Variables	UC, n (%)	CD, n (%)	p value
Extraintestinal manifestations	164 (15.57)	86 (27.74)	< 0.01
IBD-associated arthritis	81 (7.69)	35 (11.29)	0.046
Aphthous ulcer	28 (2.66)	41 (13.23)	< 0.01
Ocular manifestations	20 (1.90)	8 (2.58)	0.457
Erythema nodosum	12 (1.14)	5 (1.61)	0.509
Pyoderma gangrenosum	6 (0.57)	3 (0.96)	0.447
Nonalcoholic fatty liver disease	29 (2.75)	13 (4.19)	0.197
Cholelithiasis	28 (2.66)	14 (4.52)	0.096
Primary sclerosing cholangitis	8 (0.76)	1 (0.32)	0.404
Thromboembolism	4 (0.38)	1 (0.32)	0.883
Perianal disease	59 (5.60)	55 (17.74)	< 0.01

• Abbreviations: CD, Crohn's disease; IBD, inflammatory bowel disease; UC, ulcerative colitis.

3.2.4 Managements of IBD Patients

The 5-aminosalicylate (5-ASA) is the mainstay therapy for UC, with about 74.17% of patients receiving this treatment. Corticosteroids were the first choice for moderate-to-severe UC. About 37 (19.4%) moderate UC and 99 (33.0%) severe UC patients were prescribed for induction. The proportion of patients who received biologics or immunosuppressants was low, with a usage rate of 4.46% (47/1053) for biologics and 3.42% (36/1053) for immunosuppressants. Of the patients using immunosuppressants, 80% had EIMs or other concomitant rheumatic diseases. In addition to medication, FMT was performed in 191 patients (18.14%) as a novel alternative therapy. The rate of surgical intervention was 8.93%, and about one-third of the cases were emergency surgery. The types of surgery included colectomy, temporary enterostomy, adhesiolysis, intestinal repair, and perianal surgery.

The usage of biologics and immunosuppressants was higher in CD patients than that in UC. About 12.58% of patients received biologics, and 7.74% received immunosuppressants. The utilization rate of 5-ASA was 63.87%, which was lower than UC. Corticosteroids showed a similar usage rate in CD and UC. Table 5 presents the management patterns for UC and CD.

Table 5. Management strategy for IBD patients.

Medications	UC (n = 1053) (%)	CD (n = 310) (%)	p value
5-ASA	781 (74.17)	198 (63.87)	< 0.01
Corticosteroids	170 (16.14)	43 (13.87)	0.333
Immunosuppressants	36 (3.42)	24 (7.74)	< 0.01
Thiopurine	10 (0.95)	7 (2.26)
Methotrexate	6 (0.57)	1 (0.32)
Thalidomide	20 (1.90)	16 (5.16)
Biologics immunomodulators	47 (4.46)	39 (12.58)	< 0.01
Infliximab	33 (3.13)	34 (10.97)
Adalimumab	12 (1.14)	5 (1.61)
Vedolizumab	2 (0.19)	0
Probiotics	361 (34.28)	87 (28.06)	0.04
FMT	191 (18.14)	12 (3.87)	< 0.01
Surgery	94 (8.93)	80 (25.81)	< 0.01
Colectomy	66 (6.27)	65 (20.97)
Enterostomy	3 (0.28)	17 (5.48)
Adhesiolysis	6 (0.57)	14 (4.52)
Intestinal repair	4 (0.38)	5 (1.61)

• Abbreviations: 5-ASA, 5-aminosalicylate; CD, Crohn's disease; IBD, inflammatory bowel disease; UC, ulcerative colitis.

3.2.5 Complications and Outcomes

The complications of UC included toxic megacolon, intestinal obstruction, intestinal perforation, massive hemorrhage, and IBD-associated dysplasia. The incidence of complications was 10.1% (99/978). UC-related dysplasia was present in 53 cases (5.03%), and the rate of UC-associated colorectal cancer (UC-CRC) was 3.04% (32/1053). Among the 28 CRC cases, more than 60% were rectal cancer, and moderately differentiated adenocarcinoma was the major pathological type (Table 6).

Table 6. Complications of IBD.

Variables	UC, n (%)	CD, n (%)	p value
Intestinal obstruction	36 (3.42)	21 (6.77)	< 0.01
Intestinal perforation	13 (1.23)	7 (2.26)	0.188
Severe bleeding	22 (2.09)	—
Toxic megacolon	10 (0.95)	—
IBD-related dysplasia	62 (5.89)	8 (2.58)	0.02
Low-grade dysplasia	22	3
High-grade dysplasia	9	1
Rectal cancer	17	2
Sigmoid colon cancer	8	0
Transverse colon cancer	1	0
Ascending colon cancer	4	2
Appendiceal cancer	2	0

• Abbreviations: CD, Crohn's disease; IBD, inflammatory bowel disease; UC, ulcerative colitis.

For CD patients, intestinal obstruction was found in 6.77% of patients, and the occurrence rate was higher than UC (p < 0.01). Intestinal perforation occurred in 2.26% of patients. The complication rate of IBD-related dysplasia was 2.58%, which is lower than UC (p < 0.05).

4 Discussion

To facilitate the integrated analysis of IBD clinical information between China and the West, we developed a new IBD database. According to the Chinese IBD guidelines and consensus [11-16], some clinical items have been added on the basis of the Penn IBD database to gain a more comprehensive IBD database. The main strength of the new IBD database lies in its combination of the Eastern and Western perspectives on the disease. Although there is currently a lack of more Western data for comparative analysis, the database entries provide some insights. For example, the history of narcotic pain medications use in Western populations is not reflected in the data from the Chinese population. The 301 IBD database contains elaborate clinical characteristics for IBD, which is suitable for both inpatients and outpatients. The new database also gives more attention on overall assessment and includes more laboratory tests, examinations, and outcome records. Such data are difficult to obtain in cohorts and are often not sufficient to cover both including a large population and a large number of longitudinal measurements. In addition, the 301 IBD database incorporates all items from the Penn IBD database, making it suitable for East-West comparative studies.

The 301 IBD database can make important contributions to three area of research. First, this database can reflect the different concerns of Eastern and Western clinical practice in IBD. Virus infection is a common issue of IBD patients and significantly correlated with disease activity [18, 19]. In addition to hepatitis virus and tuberculosis, China is also concerned about coinfection with EBV, CMV, and other opportunistic pathogens. We found that the Penn IBD database contained more novel biologics, but no relevant use in Chinese patients. In the latest Chinese guideline of IBD, inflixmab, vedolizumab, and ustekinumab were recommended for the induction of remission. However, the indications of certolizumab, golimumab, and tofacitinib were not included in Chinese guidelines [20, 21]. The majority of IBD-associated dysplasia in Chinese patients was colorectal cancer, while cervical dysplasia and skin cancer had low attention and few data. Evidence had shown that there was no significant causal relationship between IBD and skin or gynecologic cancers in the East Asian population [22]. However, significant associations between CD and ovarian cancer (odds ratio [OR] = 0.898, 95% CI = 0.844–0.955) and between UC and nonmelanoma skin cancer (OR = 1.002, 95% CI = 1.000–1.004, p = 0.019) were observed in the European population [22]. Besides of hepatitis virus and Mycobacterium tuberculosis, more opportunistic infection items like EBV, CMV, and bacterial agents are added in the 301 IBD database. Vigilance is needed when monitoring for opportunistic infections since patients with IBD are at increased risk when treated with immunosuppressants and biologics [23]. More items of UC-associated complications are added to the 301 IBD database, including toxic megacolon, intestinal obstruction, and intestinal perforation. In China, UC was more common than CD [24, 25], so the complications and outcomes of UC were included in the 301 IBD database.

Second, the 301 IBD database can improve the clinician's understanding of IBD disease characteristics, especially for less experienced physicians. Inadequate disease awareness in physicians is one of the diagnostic challenges for disease management, especially in resource-limited regions [26]. IBD providers’ knowledge needs to be improved in the areas of therapeutic decision-making and treating IBD in special subsets [27]. IBD database can be a cost-effective tool to address this issue that provides diagnostic highlights and management focus to primary care physicians [28].

Third, the 301 IBD database is a favorable tool for clinical analysis of diseases. To verify the utility of the new IBD database in practice, we used the database to record the clinical information of IBD inpatients over 16 years at PLAGH from 2008. In mainland China, IBD is becoming one of the more common digestive diseases encountered in the clinic. The increasing trend of hospitalization in our results and in other studies supports this point, especially UC [29, 30]. We observed a predominance of male patients with UC and CD, and the male predominance of CD was more obvious, which is consistent with several Chinese reports but different from a similar sex distribution in Western countries [31-34]. Previous studies have reported a male-predominant pattern of IBD in Asia, suggesting that environmental and geographical factors could have a major role in sex-based differences in disease epidemiology [35]. Additionally, there is evidence that the effect of hormone signaling, differentially regulated gene transcription, gut microbiota, and immune responses could all play pathophysiological roles in the observed sex differences in IBD epidemiological patterns [36, 37]. The age-specific distribution of onset age showed a unimodal pattern with the peak range of 21–30 years in males and 31–50 years in females. Our results differed some developed countries which demonstrated a bimodal pattern of onset age, with the first onset peak before 40 years and a second peak year more than 60 [38, 39]. Age-specific incidence rate may be related to geographical location, economic level, or changes in diet. An epidemiology study from the Asia-Pacific region showed a unimodal pattern with peak age of diagnosis at 30–34 years [40].

In terms of disease phenotype, we found a more significant proportion of extensive colitis (55.05%) for hospitalized UC patients. Regarding disease severity, about 60% of patients had mild-to-moderate disease. When considered with outpatients, there was a similar distribution of disease locations in UC [25, 41]. The mild-to-moderate form has also been noted in Asian research [35]. For patients with CD, L1 and L3 were the common disease phenotypes and represented with nonstructuring, nonpenetrating. CD patients also showed a high rate of concomitant perianal disease. CD had a higher rate of EIMs than UC. The similar phenotypes were reported in other Asia cohorts [25, 41].

Treatment options for IBD include medication, surgery, and new clinical trials such as FMT. The 5-ASA remain the first-line choice based on Chinese guidelines which is supported by the data from our 301 IBD database analysis results, even with the availability of more advanced therapies such as biologics. The proportion of patients who received biologics or immunosuppressants was low for patients with UC, with a usage rate of 4.08% for biologics and 3.80% for immunosuppressants. Although CD patients had a higher utilization rate of biologics and immunosuppressants than UC, the total rate was still lower than Western cohorts. The costs and clinical experience may influence the usage of biological agents [42]. The high rate of adverse effects was an important concern regarding immunosuppressant therapy [31].

Asian patients tended to undergo a lower rate of surgery than those in the West. In some studies, the proportion of surgical procedures in UC patients was 1.3%–7.7% in Asia [33, 35] and approximately 15% in Western countries [43]. Consistently, our data showed a rate of 8.9% for surgery surgical intervention, and colectomy was the major surgical procedure. Instead, the surgery rate in CD patients was high, approximately one-quarter, which was higher than that of Western countries [44, 45]. Misdiagnosing, low utilization rate of biologics, and lack of awareness seem to play roles in this [45, 46].

IBD-associated dysplasia is the general problem worldwide. It is estimated that the incidence of UC-CRC varies from 0.27% to 3.09% based on a heterogeneous population [31, 47-49]. Our results revealed a higher incidence of 3.04% of UC-CRC than other pieces of Asian research, which reported a rate of 0.37%–2.70% [32, 35]. The high rate of CRC may be due to a lack of regular colonoscopy surveillance. It has been recommended that UC patients routinely receive endoscopy starting from the eighth year after diagnosis, which is a weak point in long-term disease management in China [50].

There are some limitations in this database that need to be addressed. First, this new database requires multicenter experience and refinement. Second, the current database has not been shared between multiple medical centers in China and other counties. Translation into multiple languages, as well as a web-based platform that maintain data security and information privacy, might facilitate future wider spread utilization. Certainly, future multicenter studies, including East-West comparative analyses, might lead to significant advances in our understanding of environmental, cultural, and genetic factors responsible for the worldwide increase in the incidence of IBD associated with industrialization.

In conclusion, the new database comprehensively integrates data fields of relevance in the care of patients with IBD in both Eastern and Western societies, namely China and the United States based on single institutional experiences (PLAGH and Penn). The uniform collection of IBD clinical information will facilitate the data sharing and deep mining of clinical information on IBD between the East and the West. Future implementation goals will focus on ease of using and accessing the database, continuing to promote the standardization of the IBD database worldwide, including accuracy of multiple language versions, with the ultimate vision of developing a database for future multicenter analysis involving real-world big data.

Author Contributions

Study design and management: Yunsheng Yang and Gary D. Wu. Database construction: Jingshuang Yan, Rongrong Ren, Hongzhe Lee, Ni Josie, Gang Sun, and Lihua Peng. Data entry: Jingshuang Yan, Ruqi Chang, Wanyue Dan, and Xiaohan Zhang. Data analysis: Jingshuang Yan, Rongrong Ren, Fei Pan, and Bin Yan. Writing and revising the paper: Jingshuang Yan.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.