Prediction of central venous catheter-associated deep venous thrombosis in pediatric critical care settings

Abstract

Background An increase in the incidence of central venous catheter (CVC)-associated deep venous thrombosis (CADVT) has been reported in pediatric patients over the past decade. At the same time, current screening guidelines for venous thromboembolism risk have low sensitivity for CADVT in hospitalized children. This study utilized a multimodal deep learning model to predict CADVT before it occurs. Methods Children who were admitted to intensive care units (ICUs) between December 2015 and December 2018 and with CVC placement at least 3 days were included. The variables analyzed included demographic characteristics, clinical conditions, laboratory test results, vital signs and medications. A multimodal deep learning (MMDL) model that can handle temporal data using long short-term memory (LSTM) and gated recurrent units (GRUs) was proposed for this prediction task. Four benchmark machine learning models, logistic regression (LR), random forest (RF), gradient boosting decision tree (GBDT) and a published cutting edge MMDL, were used to compare and evaluate the models with a fivefold cross-validation approach. Accuracy, recall, area under the ROC curve (AUC), and average precision (AP) were used to evaluate the discrimination of each model at three time points (24 h, 48 h and 72 h) before CADVT occurred. Brier score and Spiegelhalter’s z test were used measure the calibration of these prediction models. Results A total of 1830 patients were included in this study, and approximately 15% developed CADVT. In the CADVT prediction task, the model proposed in this paper significantly outperforms both traditional machine learning models and existing multimodal deep learning models at all 3 time points. It achieved 77% accuracy and 90% recall at 24 h before CADVT was discovered. It can be used to accurately predict the occurrence of CADVT 72 h in advance with an accuracy of greater than 75%, a recall of more than 87%, and an AUC value of 0.82. Conclusion In this study, a machine learning method was successfully established to predict CADVT in advance. These findings demonstrate that artificial intelligence (AI) could provide measures for thromboprophylaxis in a pediatric intensive care setting.

Central venous catheter-associated deep vein thrombosis in critically ill pediatric patients: risk factors, prevention, and treatment

“…Jaffray et al [15 ▪▪ ] utilized traditional statistics methods to develop their model from the CHAT Registry, which had good discriminatory accuracy with an area under the receiver operating curve (AUROC) of 0.79 [95% confidence interval (CI) 0.73–0.84]. Badheka et al [5 ▪ ] and Li et al [7 ▪ ] both utilized neural networks to create models for CADVT. These had similar accuracy to the Jaffray et al model with an AUROC of 0.73 and 0.82, respectively.…”

“…Central venous catheter (CVC) placement and ICU admission are the two most important risk factors for VTE in children [2]. Critically ill children are reported to develop CVC-associated DVT (CADVT) at rates of 2–39%, depending on the presence of symptoms and screening techniques [3,4,5 ▪ ,6,7 ▪ ]. CADVT can lead to pulmonary embolism, potentially prolonging duration of mechanical ventilation [8].…”

Ultrasound‐guided supraclavicular cannulation of left brachiocephalic versus right internal jugular vein: Comparative analysis of central venous catheter‐associated complications

“…A feared complication following CVC placement is catheter‐associated deep vein thrombosis (CADVT) which is associated with increased morbidity and mortality in pediatric patients 7 . Several studies have reported that CADVT occurs in approximately 15%–39% of pediatric patients with a CVC and the risk is particularly high in preterm infants and neonates 9,10 . It must be noted, that the inflammatory and immunosuppressive effects of cardiopulmonary bypass and major surgery may have a different impact on the occurrence of complications in such patients when compared with other conditions 5,6 .…”

“…7 Several studies have reported that CADVT occurs in approximately 15%-39% of pediatric patients with a CVC and the risk is particularly high in preterm infants and neonates. 9,10 It must be noted, that the inflammatory and immunosuppressive effects of cardiopulmonary bypass and major surgery may have a different impact on the occurrence of complications in such patients when compared with other conditions. 5,6 Furthermore, the question of the influence of the puncture site of a CVC on the development of CADVT for pediatric cardiac surgery patients has not been conclusively answered in studies.…”

Enoxaparin to Prevent Catheter-Associated Thrombosis: Effective, But Is This a Tool We Need?*

“…To date, many studies focused on defining risk factors for CADVT have been retrospective in nature. However, Li et al (10) have recently published an algorithm using machine learning that they have shown to predict the development of CADVT 72 hours in advance of the event. If this algorithm holds up to additional real-world experience, maybe we will find ourselves in the position where we can more precisely identify risk and likelihood of developing CADVT and therefore be able to prospectively address the question of treatment strategy.…”