Hemodilution during venous gas embolization improves gas exchange, withoutaltering V-A/Q or pulmonary blood flow distributions

Background: Isovolemic anemia results in improved gas exchange in rabbits w ith normal lungs but in relatively poorer gas exchange in rabbits with whol e-lung atelectasis. In the current study, the authors characterized the eff ects of hemodilution on gas exchange in a distinct model of diffuse lung in jury: venous gas embolization. Methods: Twelve anesthetized rabbits were mechanically ventilated at a fixe d rate and volume. Gas embolization was induced by continuous infusion of n itrogen via an internal jugular venous catheter, Serial hemodilution was pe rformed in six rabbits by simultaneous withdrawal of blood and infusion of an equal volume of 6% hetastarch: six rabbits were followed as controls ove r time. Measurements included hemodynamic parameters and blood gases, venti lation-perfusion ((V) over dot(A)/(Q) over dot) distribution (multiple iner t gas elimination technique), pulmonary blood now distribution (fluorescent microspheres), and expired nitric oxide (NO; chemoluminescence). Results: Venous gas embolization resulted in a decrease in partial pressure of arterial oxygen (Pa-O2) and an increase in partial pressure of arterial carbon dioxide (Pa-CO2), with markedly abnormal overall (V) over dot(A)/(Q ) over dot distribution and a predominance of high (V) over dot/(Q) over do t areas. Pulmonary blood flow distribution was markedly left-skewed, with l ow-flow areas predominating. Hematocrit decreased from 30 +/- 1% to 11 +/- 1% (mean +/- SE) with hemodilution. The alveolar-arterial P-O2 (A-aP(O2)) d ifference decreased from 375 +/- 61 mmHg at 30% hematocrit to 218 +/- 12.8 mmHg at 15% hematocrit, but increased again (301 +/- 33 mmHg) at 11% hemato crit. In contrast, the A-aP(O2) difference increased over time in the contr ol group (P < 0.05 between groups over time). Changes in Pa-O2 in both grou ps could be explained in large part by variations in intrapulmonary shunt a nd mixed venous oxygen saturation (Sv(O2)); however, the improvement in gas exchange with hemodilution was not fully explained by significant changes in (V) over dot(A)/(Q) over dot or pulmonary blood flow distributions, as q uantitated by the coefficient of variation (CV), fractal dimension, and spa tial correlation of blood flow. Expired NO increased with with gas emboliza tion but did not change significantly with time or hemodilution. Conclusions: Isovolemic hemodilution results in improved oxygen exchange in rabbits with lung injury induced by gas embolization. The mechanism for th is improvement is not clear.