Predictors of apnea test failure during brain death determination.

Background: In a recent publication (Wijdicks et al. in Neurology 71(16):1240, 2008), apnea test safety during brain death determination was evaluated at a single tertiary care center. One major conclusion was that apnea testing was safe in hemodynamically compromised patients in most circumstances and rarely aborted. Determinants of apnea test completion failure are unknown.

Methods: A-a gradients and PaO2/FiO2 ratios were calculated in the previously studied cohort. Arterial blood gas (ABG) values and systolic blood pressures (SBP) were recorded prior to apnea test initiation. Patients that completed the procedure during the declaration of brain death were compared to those whose studies were aborted. Statistical analysis was performed using Wilcoxon rank-sum and Fisher's exact tests where appropriate. Aborted apnea test risk factor assessment was by logistic regression analysis.

Results: 207 of the original 228 patients were evaluated. 10 of the 207 patients had aborted apnea tests because of hypoxemia and/or hypotension. 60% who failed the apnea test were male and were of younger age [median: 23 years vs. median: 47 years (P = 0.02)]. A-a gradient median values for aborted and completed apnea tests were 376 and 175 mmHg, respectively (P = 0.003). Neither the PaO2/FiO2 ratio (P = 0.14) nor SBP (P = 0.28) were associated with test completion failure. Acidemia preceding a carbon dioxide challenge was independently associated with test completion failure (P = 0.028).

Conclusion: Acute lung injury is common in patients undergoing brain death evaluation. Patients that failed completion of apnea testing tended to be younger, had significantly greater A-a gradients, and were more acidotic.