OBJECTIVES: The prevalence of cardiac arrest (CA) in hospitalized children is 0.77/1,000 admissions, with a higher relative incidence in children with heart disease. Despite the low prevalence, there is a significant cost burden, largely related to redundant therapies and morbidity. Independent of the complexity of the surgery, the rate of CA after pediatric cardiac surgery varies between institutions, being reported in 2.6 - 6% of patients with a significantly high mortality rate. Patients undergoing more complex surgeries represent post-operative cohorts with the highest risk for CA in the pediatric cardiac intensive care unit (PCICU). This review presents the challenges associated with clinical situational awareness in the PCICU and discuss the use of decision support tools which use predictive algorithms to mitigate risk of adverse events, like CA.

METHODS: A review was conducted using publications between 2009-2019, in the MEDLINE/PubMed databases, LILACS, Cochrane Library and EMBASE. Articles in English, Spanish or in Portuguese were included. Selection was performed by two reviewers independently, using the following descriptors: situational awareness; decision support techniques; heart arrest; monitoring, physiologic; pediatrics.

RESULTS: Eight out of 85 studies were included. Two (25.0%) were prospective: one assessed the human factor approach in using a data integration tool, and a second evaluated a computational simulator to predict events prior to clinical symptoms. Five (62.5%) reviewed non-invasive medical devices. All studies provided consensus on the importance of reliable systems that assist clinicians on proper identification, escalation, and mitigation.

CONCLUSIONS: This literature review suggests that clinical decision support tools, especially those that use high-fidelity data collected within the critical care ecosystem, contributes to data-driven decision making enabling clinicians to intervene earlier, thereby reducing the risk of adverse events as a result of delayed treatment. Further studies involving predictive analytics in the PCICU and economic evaluations should be encouraged.