Transcutaneous Carbon Dioxide Monitoring During Apnea Testing for Determination of Neurologic Death in Children: A Retrospective Case Series.

OBJECTIVES

Determination of neurologic death in children is a clinical diagnosis based on absence of neurologic function with irreversible coma and apnea. Apnea testing during determination of neurologic death assesses spontaneous respiration when PaCO2 increases to greater than or equal to 60 and greater than or equal to 20 mm Hg above pre-apneic baseline. The utility of transcutaneous carbon dioxide measurements during apnea testing in children is unknown. We seek to determine the degree of correlation between paired transcutaneous carbon dioxide and PaCO2 values during apnea testing for determination of neurologic death.

DESIGN

Single-center, retrospective case series.

SETTING

Twenty-eight bed PICU in a 259-bed, tertiary care, referral center.

PATIENTS: Children 0-18 years old undergoing determination of neurologic death between May 2017 and December 2018.

INTERVENTIONS: None.

MEASUREMENTS AND MAIN

RESULTS

Primary outcomes were paired transcutaneous carbon dioxide and PaCO2 values obtained during determination of neurologic death. Primary analyses included Pearson correlation coefficient, Bland-Altman bias and limits of agreement, and comparative statistics. Descriptive data included demographics, admission diagnoses, hemodynamics, Vasoactive Inotropic Scores, and arterial blood gas measurement. Eight children underwent 15 determination of neurologic death examinations resulting in 31 paired transcutaneous carbon dioxide and PaCO2 values for study. Transcutaneous carbon dioxide and PaCO2 correlated well (r = 0.94; p < 0.01). Bias between transcutaneous carbon dioxide and PaCO2 was -3.29 ± 7.14 mm Hg. Differences in means did not correlate with Vasoactive Inotropic Score (r = 0.2) or patient temperature (r = 0.11). Receiver operator characteristic curve of transcutaneous carbon dioxide after 3-10 minutes of apnea to discriminate positive apnea testing by the standard of PaCO2 yielded an area under the curve of 0.91 and threshold of greater than or equal to 64 mm Hg (sensitivity, 91.7%; specificity, 100%; positive predictive value, 100%; negative predictive value, 92.3%; accuracy, 95.9%).

CONCLUSIONS

During apnea testing for determination of neurologic death in children, noninvasive transcutaneous carbon dioxide monitoring demonstrated high correlation, accuracy, and minimal bias when compared with PaCO2. Further validation is required before any recommendation to replace PaCO2 with noninvasive transcutaneous carbon dioxide monitoring can be proposed. However, concurrent transcutaneous carbon dioxide data may limit unnecessary apnea time and associated hemodynamic instability or respiratory decompensation by approximating goal arterial blood sampling to document target PaCO2.