text in 
English (pdf)
Article in xml format
Article references
Automatic translation
Send this article by e-mail
Cited by SciELO 
Cited by Google 
Similars in
SciELO 
Similars in Google 
Permalink
Introduction
Inflammatory bowel disease (IBD) is a condition of unknown etiology that includes Crohn’s disease (CD), ulcerative colitis (UC), and inflammatory bowel disease unclassified (IBDU) in adults. In children, it also encompasses atypical ulcerative colitis (AUC) and colonic Crohn’s disease (CCD)1,2. IBD can manifest at any age, and its incidence is rising across all age groups3,4. The pathogenesis involves polygenic and epigenetic factors and is characterized by chronic inflammation of the gastrointestinal tract2,5.
The presentation, progression, behavior, and prognosis of pediatric-onset inflammatory bowel disease (PIBD) differ significantly from adult-onset IBD. The reclassification by the Porto Group since 2011 has established criteria that are highly informative for understanding PIBD. These criteria aim to guide classification for accurate diagnosis in children and adolescents by defining phenotypes based on clinical, endoscopic, histopathological, and radiological differences. The result is the PIBD-classes, which divides PIBD into five phenotypes: typical ulcerative colitis (TUC), atypical ulcerative colitis (AUC), Crohn’s disease (CD), colonic Crohn’s disease (CCD), and inflammatory bowel disease unclassified (IBDU)1,2.
Although the intestine is the primary target, other organs can be affected. In adults, 50% develop at least one extraintestinal manifestation (EIM) during the course of the disease, and EIMs can precede intestinal symptoms in 16% to 25% of cases6-9. In Colombia, a prevalence of 23.3% has been reported in adults, with joint involvement being the most frequent, as noted in other series6,10. Most EIMs occur in parallel with bowel disease activity and are more common in CD, especially in CCD patients11,12.
Patients with PIBD experience significantly more EIMs both at diagnosis and during follow-up13,14. In pediatrics, the prevalence of EIMs at diagnosis ranges from 6% to 29%15. In a cohort of 1649 patients, the cumulative incidence was 9% at one year, 19% at five years, and 29% at fifteen years post-diagnosis16. EIMs usually present in children older than six years and, like in adults, are more frequently associated with CD (61%) than with UC (29%)12,16,17, with 4% to 25% possibly preceding intestinal involvement11,12,15.
When PIBD is accompanied by EIMs, it follows a more aggressive clinical course in terms of progression and extent, necessitating increased use of corticosteroids, immunosuppressants, biological therapies14, and surgery, particularly in UC13. Increased EIM activity is associated with a higher relapse rate in CD15).
The pathophysiology of EIMs is not fully understood. Possible contributing factors include genetic predisposition, changes in the intestinal microbiome, and alterations in the immune response from the intestine to other organs, affecting leukocyte circulation12,18. Based on these factors, EIMs can be classified as classic when the inflammatory process occurs in distant sites (musculoskeletal, mucocutaneous, ocular, and hepatobiliary), associated when they occur with other immunologically mediated diseases (multiple sclerosis, myocarditis, psoriasis, systemic lupus erythematosus), and as complications in the case of anemia and malnutrition, as manifestations of systemic inflammation18-21.
Classic EIMs most frequently affect the joints, skin, and eyes but can also involve other organs such as the liver, lungs, and pancreas18. A common misconception is that successful therapy of intestinal inflammation is sufficient to satisfactorily treat EIMs22; generally, peripheral spondyloarthropathy, oral ulcers, episcleritis, and erythema nodosum may improve with standard treatment of intestinal inflammation, but uveitis, ankylosing spondylitis, and primary sclerosing cholangitis (PSC) progress independently of disease flares12.
In children, EIMs primarily affect the joints and skin13,14, and less frequently the eyes and liver, significantly impacting and worsening quality of life15. Although their improvement or resolution is part of the general management of PIBD, their presence should always be considered when deciding on therapy for managing intestinal inflammation8,12.
In this context, it is essential to understand the clinical, evolutionary, and prognostic characteristics of EIMs in the pediatric population, as well as their prevalence, to estimate their impact and thus offer a comprehensive approach in the therapy and follow-up of patients with PIBD.
Materials and methods
Study Type
A descriptive cross-sectional study was conducted through a retrospective review of medical records from 2007 to 2022. The manuscript was prepared according to the STROBE guidelines23.
Study Population
The study population consisted of pediatric patients with a confirmed diagnosis of IBD treated at the Fundación Hospital Pediátrico La Misericordia (HOMI), a high-complexity hospital and national referral center for pediatric gastrointestinal diseases located in Bogotá, Colombia (Figure 1).
Sample Selection and Size
Sampling was carried out using a non-probabilistic, sequential method. Due to the descriptive nature of the study, no formal sample size calculation was performed. All patients who met the inclusion criteria during the study period were included.
Diagnostic Criteria
The diagnosis of PIBD was made according to the recommendations of the European Society for Pediatric Gastroenterology, Hepatology, and Nutrition (ESPGHAN) established in 2014 through the Porto criteria and PIBD-classes. These criteria integrate the majority of evidence on recommended diagnostic methods and provide an initial approach for the practical classification of the five disease phenotypes1. This classification allows for a more precise diagnosis, appropriate therapeutic decisions, and prediction of disease progression and prognosis.
The diagnosis of EIMs was based on the medical evaluation by the pediatric gastroenterology team through physical examination and interdisciplinary clinical criteria recorded in the medical record. The classic EIMs analyzed included spondyloarthropathies (peripheral and axial), uveitis, pyoderma gangrenosum, erythema nodosum, aphthous stomatitis, skin tags (anal folds), psoriasis, and sclerosing cholangitis.
Anemia, malnutrition, and growth failure were considered complications and therefore were not included. Nutritional diagnosis was conducted using the current criteria of the World Health Organization (WHO), endorsed by the Ministry of Health and Social Protection of Colombia24. Anemia was defined according to WHO criteria (hemoglobin < 11 g/dL for children)18.
EIMs affecting the musculoskeletal system, particularly the joints, were grouped under the term spondyloarthropathies, using the diagnostic criteria developed by the Assessment in SpondyloArthritis International Society (ASAS), which distinguish between peripheral and axial manifestations; both can occur in PIBD patients, with special considerations for application to the pediatric population25,26.
Mucocutaneous involvement was classified into four groups:
Reactive: when they share the immunopathogenic mechanism but not the histopathological characteristics (erythema nodosum, pyoderma gangrenosum, Sweet syndrome, oral lesions).
Specific: when the histopathological characteristics are similar but present outside the gastrointestinal tract, as in metastatic Crohn’s disease27.
Treatment-associated: psoriasiform lesions, skin infections, and eczema are manifestations secondary to treatment complications, adverse or paradoxical reactions to medications used: thiopurines, sulfasalazine, methotrexate, and especially anti-tumor necrosis factor (anti-TNF)28.
Nonspecific: these lesions can be diverse, and their diagnosis is based on clinical characteristics and the exclusion of other dermatological disorders18,29.
All EIMs were diagnosed by the corresponding expert pediatricians in dermatology, rheumatology, ophthalmology, and hepatology.
To control for potential information and recall biases when retrospectively applying the PIBD-classes criteria in reclassifying PIBD, the following strategies were employed:
Integration into REDCap of the “IBD Classes (Catherine McEwasan Foundation)” application.
Verification and clarification of inconsistent data.
Review of data entry values for variables.
Simple imputation of missing data when required.
Confirmation of the appropriate phenotypic classification of patients by one of the investigators, who had prior training in its application.
Data Collection
Medical records of patients who met the diagnostic criteria for PIBD and EIM at the time of disease diagnosis were reviewed. Data collection was carried out through retrospective review of medical records, with variables recorded using a data collection form created in REDCap®, which was later exported to the statistical software STATA 18.0®.
Statistical Analysis
Descriptive statistics determined the frequency and percentages of categorical variables. For continuous variables, central tendency measures were determined according to their distribution (mean with standard deviation [SD], median with interquartile range [IQR]). The association between EIMs and demographic, clinical, and biochemical characteristics was evaluated using Fisher’s F for qualitative variables; for quantitative variables, the Mann-Whitney U test and Student’s T test were used according to the nature of the data distribution. A p-value < 0.05 was considered statistically significant.
Ethical Considerations
This study was approved by the ethics committee of Fundación Hospital Pediátrico La Misericordia (HOMI). Data were extracted from medical records, so there was no interaction with participants, and informed consent was not required.
Data were collected using an MS Excel format. No individual data or information that could identify any patient or their family were disclosed, in accordance with the Statutory Law on Data Protection.
Results
Demographic Features
Over a 15-year period, 71 cases of PIBD were confirmed, of which 18% (13/71) presented EIMs during the course of the disease. Among these, 62% (8/13) were male. The median age at diagnosis for the entire group was 11.4 years (IQR = 7.4-17.6). Table 1 summarizes the demographic characteristics of patients in the EIM and non-EIM groups.
Table 1 Demographic Characteristics of Patients with PIBD
| Does Patient Have EIM? | Total | p-Value | ||
|---|---|---|---|---|
| With EIM | Without EIM | |||
| Time in years | n =13 (18%) | n = 58 (82%) | n = 71 (100%) | |
| Age at symptom onset | 9.9 (SD: 6.1) | 9.07 (SD: 4.8) | 10.2 (IQR: 4.7-9.25) | 0.4 |
| Age at PIBD diagnosis | 13.2 (SD: 3.5) | 10.2 (SD: 4.4) | 11.4 (IQR: 7.4-17.6) | 0.02* |
| Sex | Total: 71 (100%) | |||
| Female | 5 (38%) | 27 (47%) | 32 (45%) | 0.4+ |
| Male | 8 (62%) | 31 (53%) | 39 (55%) | |
| Origin | Total: 58 (100%) | |||
| Urban | 8 (80%) | 43 (89%) | 51 (88%) | 0.3 |
| Rural | 2 (20%) | 5 (10%) | 7 (12%) | |
| Affiliation regime | Total: 71 (100%) | |||
| Contribution-based | 11 (85%) | 40 (69%) | 51 (72%) | 0.6 |
| Subsidized | 2 (15%) | 13 (22%) | 15 (21%) | |
| Private | 0 | 5 (9%) | 5 (7%) | |
+Fisher’s exact test. * Mann-Whitney U test. Author’s own research.
Clinical Features
The time from the onset of symptoms to diagnosis was longer in the group with EIMs (13.2 years versus 10.2 years; p = 0.02). Four patients had a personal history of autoimmune disease (two with arthritis and two with autoimmune hepatitis), while another four reported a history of psychiatric illness; among the latter, three had depression and one had attention deficit hyperactivity disorder. Regarding family history, four patients had a first-degree relative with IBD, and none belonged to the group with EIMs. Table 2 shows the most frequently reported symptoms and signs in the group with EIMs.
Table 2 Clinical Characteristics at Diagnosis
+Fisher’s exact test. * Mann-Whitney U test. Author’s own research.
Nutritional Status
Of the 71 patients, 42% (28) exhibited short stature or were at risk of short stature, 43% (29) were malnourished, and 10% (6) were overweight or obese. No significant differences were found between the groups analyzed with or without EIM.
Extraintestinal Manifestations
Among the 13 patients with EIMs, 54% (7/13) had one manifestation and 46% had two or more. These were more frequent in women (p = 0.08) (Table 3) and predominantly mucocutaneous, followed by musculoskeletal involvement. The frequency distribution of EIMs in the entire sample is presented in Figure 2.
Table 3 Types of EIMs
| Type | Frequency |
|---|---|
| Articular (spondyloarthropathy) | 8 (33%) |
| Axial (ankylosing spondylitis) | 1 (4%) |
| Peripheral | 7 (29%) |
| Mucocutaneous | 14 (58%) |
| Aphthous stomatitis | 3 (12.5%) |
| Erythema nodosum | 4 (16.6%) |
| Pyoderma gangrenosum | 3 (12.5%) |
| Psoriasis | 1 (4%) |
| Skin-tag | 4 (16.6%) |
| Hepatic | 1 (4%) |
| Sclerosing cholangitis | 1 |
| Ophthalmologic | 1 (4%) |
| Uveitis | 1 |
Author’s own research.
Table 4 shows the distribution of EIMs according to phenotype.
Table 4 Distribution of EIMs According to PIBD Phenotype
| Type of EIM | TUC (n = 10) | AUC (n = 0) | IBDU (n = 3) | CCD (n = 0) | CD (n = 12) |
|---|---|---|---|---|---|
| Articular | |||||
| Axial: sacroiliitis | 1 | 0 | 0 | 0 | 0 |
| Peripheral | 4 | 0 | 1 | 0 | 2 |
| Cutaneous | |||||
| Aphthous stomatitis | 1 | 0 | 0 | 0 | 2 |
| Erythema nodosum | 0 | 0 | 1 | 0 | 3 |
| Pyoderma gangrenosum | 1 | 0 | 0 | 0 | 2 |
| Psoriasis | 1 | 0 | 0 | 0 | 0 |
| Skin-tag | 1 | 0 | 1 | 0 | 2 |
| Hepatic | |||||
| Sclerosing cholangitis | 1 | 0 | 0 | 0 | 0 |
| Ophthalmologic | |||||
| Uveitis | 0 | 0 | 0 | 0 | 1 |
Author’s own research.
Paraclinical Tests
Among the 71 patients, half were anemic. The hemoglobin levels were significantly lower in the group with EIMs (9.8 [SD: 2.7] vs. 12.3 [SD: 2.8] g/dL; p = 0.004). Albumin levels were also lower in this group, and they exhibited higher erythrocyte sedimentation rates (ESR: 36.4 [SD: 12.3] vs. 22 [SD: 17.3] mm/h, p = 0.02) and higher platelet counts (Figure 3). Calprotectin levels were reported in 21 patients, with 17 of them showing positive results (> 250 μg/g).
Treatment
Induction treatment with biological therapy was more common in the EIM group compared to the non-EIM group (38% vs. 23%). Four of the 71 patients required surgical intervention as part of their treatment, including two colectomies with ileostomy, one drainage of an intra-abdominal abscess, and one exploratory laparotomy. None of these four patients were in the EIM group.
Discussion
This study examined the prevalence of EIMs in the largest cohort of patients with PIBD over a 15-year period in Colombia. The findings reveal that EIMs are common, with a prevalence of 18% at the time of diagnosis, which aligns with previously reported figures in the literature ranging from 6% to 29%15. Additionally, it has been reported that between 4% and 25% of EIMs could precede intestinal involvement11,12. Furthermore, the presence of MEIs is associated with greater clinical and paraclinical severity of PIBD.
The prevalence of EIMs in PIBD varies significantly across published studies, likely due to the lack of unified criteria for defining them in pediatrics, as current guidelines are primarily based on adult data11. This variability is reflected in studies reporting prevalences as high as 80%, likely an overestimation due to the inclusion of osteopenia and growth retardation, which are now considered complications of PIBD18. While the prevalence at diagnosis found in this study is similar to the reports by Jansson and colleagues (14.1%) and Duricova and colleagues (8.9%)13,15, it is notable that it is slightly higher in the present cohort, possibly due to the interdisciplinary approach.
This study also found that the time between the onset of symptoms and the age at diagnosis of PIBD was significantly longer in cases with EIMs. This delay in definitive diagnosis may suggest that the initial evaluation was directed at other organs or systems rather than the digestive system7.
EIMs often involve more than one organ or system. In this study, nearly half (6/13) of the patients with EIMs had involvement in multiple organs or systems. In fact, the likelihood of developing an EIM increases as the disease progresses and with the prior presence of an EIM11. Over the course of the disease, up to 50% of adult patients with IBD will develop at least one EIM, and this involvement can be a significant source of morbidity and even a risk factor for mortality, especially in cases such as primary sclerosing cholangitis or venous thromboembolic events18.
In the present cohort, mucocutaneous manifestations were the most common, occurring in 58% of cases. Mucocutaneous involvement has been described in 15%-20% of patients and is often determined by the shared underlying immunopathogenic mechanisms of PIBD18,29. In the pediatric population, reports of mucocutaneous involvement are scarce, but this study suggests a higher association with CD, which aligns with existing literature27. These manifestations are strongly linked to intestinal inflammatory activity and, in up to 50% of cases, appear simultaneously with the initial gastrointestinal symptoms29. Diagnosis is based on clinical characteristics and the exclusion of other dermatological disorders. Given the heterogeneity of lesions, diagnosing these manifestations can pose a challenge for pediatric gastroenterologists, emphasizing the importance of interdisciplinary evaluation, including dermatology, and considering the psychological implications and impact on quality of life18,27.
Musculoskeletal involvement (spondyloarthropathies) was the second most frequent EIM in this cohort. The association between IBD and axial or peripheral spondyloarthropathy indicates a common inflammatory pathway; genomic studies have identified shared loci that increase the risk of these diseases, involving the interleukin (IL)-12/23 signaling pathways18. Recent systematic reviews in adults have described peripheral joint involvement as the most frequent manifestation, with a prevalence of 13%, similar to data reported in adults in Colombia6,30. In this study, the prevalence is 33%, which is consistent with other reports, although the prevalence reported in pediatrics varies between 2% and 35%, highlighting a lack of standardized definitions and criteria31.
The most frequently reported signs and symptoms in these 13 patients were nocturnal diarrhea, weight loss, and pallor. Additionally, these patients exhibited significantly lower levels of hemoglobin and albumin, along with elevated ESR. These findings are considered indicators of higher disease activity and greater systemic involvement. This is consistent with multiple studies indicating a more aggressive clinical course in terms of progression and extension when PIBD is associated with EIMs19, with a higher likelihood of corticosteroid use, immunosuppressive therapy14, biological therapy, and surgical management, especially in UC13. In CD, EIMs are associated with a higher relapse rate15.
The prevalence of anemia in all PIBD cases was 50%. Although anemia was not considered an EIM, the most recent guidelines from the European Crohn’s and Colitis Organisation (ECCO) define it as a complication of the disease18. Thus, it is a fundamental aspect to evaluate and treat in patients with IBD, given its negative influence on disease progression32. Reports of anemia in adults are highly variable, ranging from 6% to 74%, likely due to the heterogeneity of the patients included in studies18. In children, the prevalence of anemia at the time of diagnosis has been described as high as 90%21,33. In this cohort, the high prevalence of anemia is particularly relevant due to the potential consequences in children during critical phases of growth and development32. Calprotectin levels do not appear to be related to EIMs, and there is no evidence to support its routine use for their identification, as it is a nonspecific marker of inflammation34.
Among the study’s limitations is the risk of information bias due to its retrospective design. Data taken only at the time of diagnosis may lead to an underestimation of cases. Over the 15-year study period, scientific advances may have led to lower initial diagnosis rates.
Regarding the COVID-19 pandemic, which occurred during the study period and caused significant global changes in life and health, its occurrence could not have been anticipated in the study design.
This study’s strength lies in being the first in Colombia to describe the frequency of EIMs in PIBD, with an interdisciplinary approach that enabled comprehensive management of these patients at a high-experience center.
Conclusions
EIMs are common in PIBD and their presence is associated with greater disease severity and progression. This study corroborates this association through the presence of symptoms and certain biochemical markers. However, these findings need to be confirmed in prospective investigations.
The diagnosis of PIBD in patients with EIMs tends to occur at older ages, possibly due to a clinical focus on systems other than the digestive system, representing an opportunity to improve early diagnosis.
In the pediatric context, it is essential to provide an interdisciplinary approach in specialized centers that comprehensively address all aspects related to the diagnosis, treatment, and follow-up of patients with PIBD associated with EIMs, to not overlook the potential impact on disease course, pharmacological treatment decisions, quality of life, and emotional well-being of patients.
