Inflammatory bowel diseases (IBD), consisting of Crohn’s disease (CD), ulcerative colitis (UC), and unclassified IBD (IBD-U), are chronic inflammatory disorders affecting the small and large intestines, with frequent extraintestinal manifestations (EIM) (1–7).

Overall, thromboembolism (TE) is a rare but potentially life-threatening event in children and adolescents. The annual incidence in children in general, excluding neonates, is reported to be 0.05–0.07 per 10,000 children. Paediatric TE includes a heterogeneous population with variable age, localisation of thrombosis, and basic comorbidities. To date, treatment is based on adult trials and expert opinion, due to a lack of evidence in children (8).

The mechanisms by which IBD causes TE are multifactorial and not fully understood (9). However, data suggest that an interplay and dysregulation of various systems lead to TE. IBD is known to involve a complexity of pro-inflammatory mediators that could contribute to a hypercoagulable state (10, 11). Furthermore, coagulation factors, including fibrinogen and fibrin products, are elevated in an acute flare of IBD, and thrombocytosis is detected in active disease states (12). In addition, lower levels of protein S and antithrombin were measured in patients with IBD (13, 14). A sole effect of inflammation seems unlikely, due to the lack of increased risk for TE in coeliac disease or rheumatoid arthritis (1).

The most frequent genetic factors in thrombophilia are prothrombin G20210A polymorphism, Factor V Leiden, and C677T polymorphism of methylenetetrahydrofolate reductase (MTHFR), less often Protein C and S or antithrombin deficiencies. When comparing these traits in adult populations with and without TE, no significant increase in prevalence rate was found (15, 16).

The risk of developing IBD is reduced in patients with inherited bleeding disorders, which could propose a link between vascular thrombosis and intestinal inflammatory processes (17). Although quantitative and qualitative alterations of multiple components of cascade factors, fibrinolytic system, and anticoagulants were discovered, they cannot fully explain the increased TE risk. However, a complex mechanism of amplification of inflammation and coagulation involving platelets and endothelial cells is proposed in patients with IBD (11).

Relevant acquired risk factors include fluid depletion, immobilisation, (emergency) surgery, central venous lines, steroid therapy, oral contraceptives, infectious complications, (cigarette) smoking, and vitamin deficiency (folate, vitamin B6, and B12) with hyperhomocysteinemia (11, 18, 19). A registry-based study on the role of TE in a large European cohort of paediatric patients with IBD gesellschaft für pädiatrische gastroenterologie und ernährung – large paediatric patient registry (CEDATA GPGE) is overdue (20).

This analysis aimed therefore to retrospectively assess TE prevalence in children and adolescents enrolled in the CEDATA GPGE database from Germany and Austria to gain further insights into TE risk in the paediatric population diagnosed with IBD.

To date, the risk of TE in paediatric IBD has been addressed in three studies only, namely of US, Canadian, and Danish cohorts, respectively, and was found to be much higher than in paediatric patients without IBD (5, 21, 22). One reason for the scarcity of data on this particular complication may be that TE only very rarely occurs in the paediatric population, with an annual incidence of 0.7–1.4 per 100,000 children or adolescents and 5.3/10,000 hospitalisations (23, 24). Based on US in-patient data by means of an aggregated, stratified random sample across five time periods, Nylund et al. (25) reported the following cumulative outcome data: The absolute risk for any TE was low, with 50.4/10,000 hospitalizations in children without IBD compared with 119.8 in CD and 101.7 in UC (RR 2.37[2.16–2.61], 1.99[1.51–2.64]), respectively.

Inflammatory bowel disease per se is an acknowledged risk factor with pathogenic and clinical specifics accountable for increased TE events, as shown in adult patients with an Odds Ratio of 1.85 [1.7–2.01] in UC and 1.48 [1.35–11.62] in CD, resulting in excess mortality of 2.1-fold compared with patients without IBD (26). The risk seems to multiply during moderate to severe flares with a hazard ratio of 4–8 (27), wherein patients with IBD in active disease sharing, for example, thrombocytosis as a significant trigger of TE (28), and especially those < 40 years having a higher risk of cardiovascular disease (29).

With regard to IBD sub-entities, a paediatric review revealed UC as significantly more frequently associated with TE than CD (OR 3.7 [1.8–7.6]) (23). This was questioned by Barclay et al. in 2010 (3), who compared reports published before and after 2000 and suggested a secular trend toward TE in CD. Data that are more recent do not show any significant trend toward higher rates for TE in UC when compared with IBD in general (22), which is, however in contrast to our findings.

When comparing our study group with the Danish cohort (21), TE incidences vary considerably by age, e.g., in the Danish study, this rate was 2/10,000 person-years in healthy controls < 20 years, 9.8 in UC, 7.8 in CD, and 8.9 in IBD. The TE incidence in our slightly younger study group (<18 years) was higher, with 14/10,000 person-years (5.5 in CD, 29.1 in UC). A further explanation for this higher rate could be the very nature of CEDATA GPGE, as mainly patients from tertiary or large paediatric IBD centres caring for more severe cases are included in this registry. In the Canadian cohort of 3,593 patients with IBD, TE incidence was 81.16 (1 year) and 31.18 (5 years) per 10,000 person-years. TE incidences were particularly high in UC patients, with a higher risk for deep vein thrombosis and pulmonary embolism than in healthy controls (the latter observation was also true for CD patients). Overall, this cohort’s TE incidences were approximately two and three times higher than those of our and the Danish studies, respectively. In adult populations studied, the relatively younger patients were at higher risk of experiencing TE and had a higher risk of recurrence (30, 31). When comparing age at TE, we did not find any relevant differences between patients affected, but the share of those who had been diagnosed with IBD under the age of 10 years was significantly higher.

In a review from 2011 (23), Lazzerini et al. found that 38% of patients affected were younger than 12 years and 11% younger than 6 years, a proportion which is confirmed by our observations. Regarding latency until TE, the authors reported a median of almost 2 years with a wide range (1 month–8 years), which is again in-keeping with our findings (median 2 years, range 0–10 years). As was shown in our study, data from the Canadian cohort demonstrated a higher incidence of TE in the first year after IBD diagnosis, obviously due to high disease activity and induction therapy (e.g., corticosteroids). While, in some contrast to this observation, among adult patients up to 30% experienced TE during well-controlled disease (27, 32), TE in paediatric IBD was in fact mostly associated with active disease, e.g., in some 82.8% of children described (25). This correlates excellently with our own findings, since all but one patient showed clinically active inflammation at the time of TE (90%), although only one CD patient had a PCDAI > 30 and 2 UC patients had a PUCAI > 45, i.e., the majority of our patients showed moderate disease activity.

Disease extension may also play a role regarding the risk for TE, as was shown in adult and paediatric studies. Not only complications like fistulas or strictures are associated with an increased risk, but especially pancolonic involvement in UC and colonic involvement in CD (26), a notion further substantiated by our findings, since all patients affected had colonic involvement. There was no statistically significant difference in rates between girls and boys (29).

The most common sites for relevant TE are deep venous thrombosis of the lower limb, pulmonary embolism, and superior sagittal sinus thrombosis with an incidence of venous TE of some 2 per 1,000 persons per year in Western populations (33, 34). In a paediatric cohort, 54.3% suffered from cerebral, 26% from limb, 13% from abdominal, and 3.3% from retinal and pulmonary involvement, respectively (23). In our cohort, the distribution was somewhat similar, with half of the patients experiencing cerebral, and limb involvement, as well as extended large intracardial thrombosis, but neither retinal nor pulmonary embolism.

Recurrence is reported in 11% during the same hospitalisation, while 10% suffer later from a second episode. In our cohort, 1/12 patients experienced a second TE 1 week after myocardial infarction, reflecting the percentage indicated in the literature.

Fortunately, in our cohort there was no mortality, so we cannot comment on mortality rates that are reported to be significantly higher in patients with IBD (OR 2.5 [1.83–3.43]) (26).

In keeping with previous reports on a rise in both TE and IBD incidences, with a higher risk of morbidity and even mortality in younger patients with IBD (morbidity OR 3.22–4.3/mortality OR 2.1) (3, 31), the findings of our study underline that these patients have a substantially increased risk for TE. Consequently, when managing paediatric IBD, an active search for TE risk factors must be undertaken and preventive measures taken, particularly for those hospitalised. Initiation of pharmacological thromboprophylaxis remains a challenge as there are no published trials for the efficacy and safety of primary pharmacological thromboprophylaxis in paediatric IBD. Further studies are needed to clarify the risks and benefits of thromboprophylaxis in this patient group and to identify the most important risk constellations for developing TE.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

AB, AH, and JD contributed to the conception and design of the study and the CEDATA-GPGE study group acquired the data. SW, CB, MC, ML, and TL provided the additional insights and information. CB, FR, JD, and AH analyzed and interpreted the data. JD, FR, and AH drafted the article and revised it critically for important intellectual content. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.