A retrospective study to determine the relationship between early microporous membrane oxygenator (MMO) failure and blood pressure at the MMO outlet (Pmo) was conducted using data collected with 19 dogs (22 +/- 1 kg, mean +/- SEM) undergoing routine normothermic cardiopulmonary bypass. Because gas flow was maintained at a high level, it could not be used to control CO2 exchange. Instead, blood PCO2 was controlled by adding CO2 to the sweep gas. Blood PO2 was controlled as suggested by the manufacturer, by adjusting the %O2 in the gas phase (g). Blood flow was 2575 +/- 54 ml/min; Pmo ranged from 173 to 790 mm Hg; and hematocrit was 33 +/- 1%. O2 exchange was calculated from blood gas parameters. Changes in the diffusion potential of O2 (delta PO2) and CO2 (delta PCO2) and MMO performance (P, taken as oxygen exchange normalized to a diffusion potential of 100 mm Hg) indicated MMO failure. Initial values, taken within 60 min of bypass initiation, were compared to final values taken at 226 +/- 9 min of bypass. Despite higher final delta PO2 (411 +/- 9 vs. 538 +/- 19 mm Hg, p less than 0.0001 paired t-test) and delta PCO2 (18.6 +/- 2.4 vs. 30.5 +/- 4.7 mm Hg, p less than 0.0017), arterial blood PO2 decreased (159 +/- 15 to 89 +/- 6 mm Hg, p less than 0.0005) and PCO2 increased (36.4 +/- 1.5 to 46.1 +/- 3.0 mm Hg, p less than 0.0039), and the performance decreased [24.5 +/- 1.1 to 20.1 +/- 0.7 (ml/min)/(100 mm Hg), p less than 0.0001].(ABSTRACT TRUNCATED AT 250 WORDS)