Standard external cardiopulmonary resuscitation (SECPR) produces high cerebral venous and intracranial pressure peaks, low cerebral perfusion pressure, and low cerebral blood flow (CBF). Cerebral viability seems to require 20% of normal CBF, which SECPR cannot reliably generate. We tested the hypothesis that SECPR can produce adequate CBF if started immediately, but not if started after a long period of cardiac arrest (no flow, stasis). Cardiac arrest times of 1, 3, 5, 7 and 9 min were studied in rabbits. We measured unifocal cortical CBF with H2 clearance curves after saturation with H2 10%, O2 50% and N2O 40% by intermittent positive-pressure ventilation (IPPV). Measurements were made during spontaneous circulation (control condition), and then after resaturation immediately before induction of asystole by KCl i.v., and H2 clearance starting at end of arrest time during SECPR-basic life support with IPPV 100% and manual chest compressions (120/min) during asystole. Control cortical CBF was 30-40 ml/100 g brain per min. During asystole and SECPR, CBF greater than 20% normal was achieved only after no-flow of 1 min. After longer arrest (no-flow) times, CBF was less than 20% normal. Values were near zero after 7 and 9 min of cardiac arrest. Decrease in mean arterial pressures (MAP) produced by SECPR during asystole paralleled CBF values. Thus, the longer the preceding period of stasis, the lower the MAP and CBF generated by SECPR without epinephrine. This effect may be the result of anoxia-induced vasoparalysis and stasis-induced increased blood viscosity.