Objective:
To derive and validate a score that correlates with an objective measurement of a child’s effort of breathing (EOB).
Design:
Secondary analysis of a previously conducted observational study.
Setting:
The pediatric and cardiothoracic Intensive Care Units of a quaternary-care children’s hospital.
Patients:
Patients more than 37 weeks gestational age to age 18 years who were undergoing extubation.
Interventions:
EOB was measured in patients following extubation using esophageal manometry to calculate pressure rate product (PRP). Simultaneously, members of a multidisciplinary team (nurse, physician and respiratory therapist) assessed respiratory function using a previously validated tool. Elements of the tool that were significantly associated with PRP in univariate analysis were identified and included in a multivariate model. An EOB score was derived from the results of the model using data from half of the subjects (Derivation cohort), and then validated using data from the remaining subjects (Validation cohort) by calculating the area under the receiver-operator characteristic curve (ROC AUC) for PRP >90th percentile and for the need for reintubation.
Measurements and Main Results:
Among 409 subjects, the median age was 5 months and nearly half were cardiac surgery patients (49.1%). Retractions, stridor and pulsus paradoxus were included in the Simple Score. ROC AUC for PRP >90th percentile was 0.8359 (95% CI: 0.7996–0.8722) in the Derivation cohort and 0.7930 (0.7524–0.8337) in the Validation cohort. ROC AUC for reintubation was 0.7280 (0.6807–0.7752) when all scores were analyzed individually, and was 0.7548 (0.6644–0.8452) if scores from three clinicians from different disciplines were summated. Results were similar regardless of provider discipline or training.
Conclusions:
A scoring system was derived and validated, performed acceptably to predict increased EOB or need for advanced respiratory support, and may function best when used by a team.