Reliability and validity of the Pediatric Intensity Level of Therapy (PILOT) scale: a measure of the use of intracranial pressure-directed therapies

Abstract

Objective: To test the reliability and validity of the Pediatric Intensity Level of Therapy (PILOT) scale, a novel measure of overall therapeutic effort directed at controlling intracranial pressure (ICP) in the setting of severe (Glasgow Coma Scale of <or= 8) pediatric traumatic brain injury (TBI).

Design: Case-control study via retrospective review of medical records.

Setting: Tertiary-care, university-based children's hospital intensive care unit.

Patients: Randomly selected patients <or=18 yrs old admitted to the intensive care unit in 2002-2003 with severe TBI (cases: group 1, n = 27), mild-moderate TBI (control: group 2, n = 30), extracranial trauma (control: group 3, n = 29), or nontraumatic illnesses (control: group 4, n = 27).

Interventions: None.

Measurements and main results: A 38-point scale was developed to quantify daily ICP-directed therapeutic effort. All currently recommended therapies are represented. Demographic and physiologic data were collected on all patients. A total of 24 of 27 patients with severe TBI received ICP-directed therapy; three did not because of judgments of futility. No control patients received ICP-directed therapy. The PILOT scale score was assessed for the first 7 days posttrauma or postadmission. Interrater reliability was 0.91 (intraclass correlation coefficient) and intrarater reliability was 0.94. The highest PILOT scale scores were in patients with severe TBI (11.7 +/- 5.6 vs. 1.3 +/- 1.7 vs. 2.0 +/- 2.1 vs. 1.9 +/- 1.8 for groups 1, 2, 3 and 4, respectively [mean +/- sd]; p < .001 by analysis of variance/Bonferroni). Patients with severe TBI who received ICP-directed therapy had higher PILOT scale scores (12.6 +/- 5.3 vs. 5.0 +/- 3.0, p = .001) than those who did not. Pearson's correlation coefficients of mean PILOT scale scores with measures of injury severity, outcome, and ICP were as follows: Glasgow Coma Scales score, -0.73 (p < .001); overall Injury Severity Score, 0.37 (p < .001); Injury Severity Score (head component only), 0.53 (p < .001); 6-month Glasgow Outcome Scale, -0.26 (p = .006); ICP burden (hours per day with ICP above treatment threshold), 0.59 (p = .002); and mean ICP, 0.41 (p = .044).

Conclusions: The PILOT scale score can be obtained retrospectively and has good reliability. It can discriminate patients receiving ICP-directed therapy, even among patients with severe TBI, and correlates with measures of injury severity, outcome, and ICP in an expected way. Thus, it seems to be a valid measure of the use of ICP-directed therapy, although prospective, multiple-center validation is recommended.