Prediction of Neurological Outcome of Children After a Traumatic Brain Injury Based on an Integrated Predictive Model

St. Justine's Hospital

Status: Unknown

Conditions: Traumatic Brain Injury

Treatments: Diagnostic Test: Prognostication model

Study type: Interventional

Funder types: Other

Identifiers

NCT04157634

HSJ - 2020-2526

Details and patient eligibility

About

This study aims to develop a integrated predictive model based on serum biomarkers, HRV, and an innovative computerized classifier output, to predict the patient long term neurological outcome after a moderate or severe TBI in children.

Full description

Traumatic brain injury (TBI) is a major cause of morbidity and mortality in children. Most children with moderate and severe TBI have long term sequelae including neurological deficit, cognitive impairment and behavioural disorders. In the acute care setting, neither clinicians nor researchers are able to adequately predict the long term outcome of children with TBI, consequently limiting their ability to tailor medical care, rehabilitation and support services. Improving our understanding of a TBI patient's exact cerebral status and prognosis is a critical step toward optimized and personalized patient management. In this research study, an innovative and integrated model will be developed to improve the prognostication in the early phase of a TBI. This model will combine key clinical variables commonly collected in the acute care setting and combine these with cutting-edge empirical measures: 1) biomarkers; 2) a new physiological monitoring based on heart-rate variability (HRV) to assess the integrity of the autonomic system; and 3) a computerized classification tool developed using the concept of artificial intelligence to continuously categorize the patient's cerebral status.

Enrollment: 70 estimated patients

Sex: All

Ages: Under 18 years old

Volunteers: No Healthy Volunteers

Inclusion criteria

Children 1) <18 years; 2) admitted to the Paediatric Intensive Care Unit at the Centre Hospitalier Universitaire Ste-Justine; 3) moderate (Glasgow Coma Scale (GCS): 9-12) or severe TBI (GCS ≤8), assessed after initial resuscitation; 4) admitted to the Paediatric Intensive Care Unit less than 24 hours after the initial TBI and 5) written parental consent. Patients with either inflicted or accidental TBI will be included as they characterize different but important population, our model should be valuable for both.

Exclusion criteria

Children 1) suspected of being brain death at the time of Paediatric Intensive Care Unit entry (GCS 3 and loss of all brain stem reflexes); 2) with a pacemaker (HRV monitoring unreliable); and/or 3) patients or parents who do not speak or read English or French.

Trial design

Primary purpose: Other

Allocation: N/A

Interventional model: Single Group Assignment

Masking: None (Open label)

---

70 participants in 1 patient group

Prognostication model

Experimental group

Description:

In a prospective cohort of children hospitalized in a PICU, development of a model based on biomarkers, HRV, and a computerized classifier output, to predict long-term neurological outcome after a moderate or severe TBI in children aged 0 to 18 years.

Treatment:

Diagnostic Test: Prognostication model