Discriminant properties of the Behavioral Pain Scale for assessment of procedural pain-related distress in ventilated children

Objectives Children hospitalized in a pediatric intensive care unit (PICU) are frequently exposed to distressing and painful medical procedures and interventions. There is a lack of clinical scales to measure procedural pain-related distress in ventilated children. The Behavioral Pain Scale (BPS) was initially developed to detect procedural pain in critically ill adults. This study aims to assess the BPS's discriminant properties for measuring procedural pain-related distress in ventilated pediatric patients incorporating two instruments validated for pediatric patients. Methods This prospective exploratory study was performed with ventilated children admitted to the interdisciplinary 14-bed PICU of the University Children's Hospital, University Medical Center Hamburg-Eppendorf, Germany. The nurse in charge and an independent observer simultaneously assessed the patients using German versions of the BPS, the COMFORT-B scale (CBS), and the modified Face, Legs, Activity, Cry, Consolability (mFLACC) scale immediately before and during endotracheal suctioning. Results We analyzed 170 parallel assessments in n=34 ventilated children. Patients were (mean +/- SD) 9.5 +/- 4.8 years old. Internal consistency for the BPS was excellent (alpha=0.93). We found a high rater agreement for all clinical scales (BPS: k=0.73, CBS: k=0.80, mFLACC: k=0.71). Strong correlations were identified between BPS and CBS (r=0.89) and BPS and mFLACC (r=0.79). The BPS cutoff values showed likewise excellent results (area under the curve CBS >16: 0.97; mFLACC >2: 0.91). Conclusions In our population of ventilated children, the BPS was well suited to detect procedural pain-related distress compared with two validated pain scales. Further extensive validation studies should follow to support our findings.