The apnea test for the determination of brain death.

By conventional criteria, an apneic patient's PaCO2 must be greater than 60 mm Hg before apnea can be attributed to brain death. The rate of a PaCO2 increase in the apneic patient traditionally has been thought to be in the range of 3 mm Hg/min. In order to assess the validity of these data and the validity of the "apnea test" for determination of brain death, the results of this test were reviewed in 20 patients. In all patients, arterial blood samples were drawn for blood gas measurements every 2 minutes following the cessation of volume ventilation (with an oxygen cannula at 6 liters O2/min passed into the tracheobronchial tree). The rate of PaCO2 increase was noted to be very erratic. The average rate of rise was 3.7 +/- 2.3 mm Hg/min (+/- standard deviation). This, however, varied from 0.5 to 10.5 mm Hg/min and was not predictable from the variables evaluated. The rate of PaCO2 increase was noted to decline throughout the duration of the test. This ranged from 3.9 +/- 1.2 mm Hg/min (for patients with baseline PaCO2 less than or equal to 30 mm Hg) and 4.5 +/- 1.9 mm Hg/min (for patients with baseline PaCO2 greater than or equal to 30 mm Hg) in the first 4 minutes of the test to an average of 0.92 mm Hg/min for patients with test lasted longer than 12 minutes. These unpredictable results might be related to CO2 washout, atelectasis, cardiac ventilations, or other yet-undefined parameters. The nonlinear relationship between rate of PaCO2 increase and time following onset of apnea resulted in the test being prolonged in several patients. In these patients, the PaCO2 approached 60 mm Hg in an asymptotic fashion. These lengthy tests could have been avoided by utilizing a standardized apnea test with a baseline PaCO2 of 40 mm Hg or greater. The observation that a high baseline PaCO2 greatly augments the efficiency and safety of the test allows criteria that have previously been based on conjecture to be documented and applied clinically. A standardized apnea test, utilizing these principles, may satisfy many of the criticisms regarding brain-death testing that have been raised by neurologists, neurosurgeons, and transplant surgeons.