Inert gas analysis of ventilation-perfusion matching during hemodialysis.

The mechanism of hypoxemia during hemodialysis was investigated by the multiple inert gas elimination technique in anesthetized, paralyzed, mechanically ventilated dogs. Profound leukopenia occurred in the first hour of a 2-h hemodialysis with a cuprophan membrane and dialysate that contained acetate. Arterial partial pressure of O2 and CO2 and oxygen consumption remained unchanged during dialysis. Pulmonary carbon dioxide elimination and lung respiratory exchange ratio decreased with the initiation of dialysis, remained depressed throughout the duration of dialysis, and returned to predialysis levels after the cessation of dialysis. Cardiac output diminished during dialysis but did not return to base-line levels after dialysis. Multiple indices calculated from inert gas analysis revealed no ventilation-perfusion mismatching during dialysis. The shunt and perfusion to regions of low alveolar ventilation-to-perfusion ratio (VA/Q) were unchanged during dialysis. There was no change in the mean or standard deviation of the profile of the percentage of total perfusion to regions of the lung that had VA/Q near 1.0; nor was there any increase in the directly calculated arterial-alveolar partial pressure differences for the inert gases during dialysis. Dead space became mildly elevated during dialysis. These results show that during dialysis with controlled ventilation there is no ventilation-perfusion mismatching that leads to hypoxemia. During spontaneous ventilation any hypoxemia must occur due to hypoventilation secondary to the CO2 exchange by the dialyzer and subsequent reduction in pulmonary CO2 exchange.