Incidence and Predictors of Venous Thromboembolism in Pediatric Intestinal Failure

Background: Venous thromboembolism (VTE) is increasingly being recognized as a complication in patients with pediatric intestinal failure (PIF) who have long-term central venous catheters (CVC). The aim of this study is to quantify the rate of VTE in patients with PIF and identify associated risk factors.

Methods: A retrospective cohort analysis was performed of all children ( <21 years of age) with severe PIF (>90 consecutive days of PN) secondary to short bowel syndrome who were treated from 2014 to 2021 at an interdisciplinary intestinal rehabilitation program. The primary outcome was the incidence of VTE, defined as any of the following identified by available imaging studies: acute deep venous thrombosis (DVT), pulmonary embolism (PE), or chronic thrombosis/occlusion of any central veins. Univariate analyses were performed to identify clinical characteristics associated with VTE. Multivariable regression was then performed to evaluate independent predictors of VTE; variables with p<0.1 on univariate analysis were included in the model.

Results: A total of 263 patients (40.3% female) were included. Median age was 7.3 years (IQR 4.2, 11.0) at most recent follow up. Seventy-four patients (28.1%) developed 114 VTEs, with an overall rate of 0.32 VTEs/1000 catheter-days. Of the patients with VTE, 23 (31.1%) had more than one thromboembolic event. Of all 114 VTEs, 30 (26.3%) extended into more than one contiguous venous segment and 61 (53.5%) were diagnosed as acute thromboses. The internal jugular and subclavian veins were the most common sites of thrombosis (31.5% each). On univariate analysis, number of catheter days [1227 (VTE) vs. 860 (no VTE); p=0.039], median number of catheters (6 vs. 4; p<0.001), and history of central line associated blood stream infection (CLABSI) (68.9% vs. 50.8%; p=0.009) were significantly associated with VTE diagnosis. There was no significant difference in gestational age, residual bowel length, underlying PIF diagnosis, total PN days, or documented liver disease between the two groups (Table 1). On multivariable logistic regression, number of catheters was the only statistically significant independent risk factor for development of VTE [aOR 1.2 (95%CI 1.09, 1.32), Table 2].

Conclusion: In this retrospective study, a striking 28.1% of patients with severe PIF developed VTE. Number of central venous catheters was the only identified independent predictor of VTE diagnosis. These data confirm VTE as an important and common complication of PIF and highlight the need for prospective studies to identify the true incidence of VTE and assess the potential benefit of interventions to decrease catheter loss, routine surveillance imaging and prophylactic anticoagulation.