SIGNIFICANT REDUCTION IN MINUTE VENTILATION AND PEAK INSPIRATORY PRESSURES WITH ARTERIOVENOUS CO2 REMOVAL DURING SEVERE RESPIRATORY-FAILURE

Objectives: To quantify CO2 removal using an extracorporeal low-resist ance membrane gas exchanger placed in an arteriovenous shunt and evalu ate its effects on the reduction of ventilatory volumes and airway pre ssures during severe respiratory failure induced by smoke inhalation i njury. Design: Prospective study. Setting: Research laboratory. Subjec ts: Adult female sheep (n = 5). Interventions: Animals were instrument ed with femoral and pulmonary arterial catheters and underwent an LD50 cotton smoke inhalation injury via a tracheostomy under halothane ane sthesia. Twenty-four hours after smoke inhalation injury, the animals were reanesthetized and systemically heparinized for cannulation of th e left carotid artery and common jugular vein to construct a simple ar teriovenous shunt. A membrane gas exchanger was interposed within the arteriovenous shunt, and blood flow produced by the arteriovenous pres sure gradient was unrestricted at the time of complete recovery from a nesthesia. CO2 removal by the gas exchanger was measured as the produc t of the sweep gas flow (FIO2 of 1.0 at 2.5 to 3.0 L/min) and the exha ust CO2 content measured with an inline capnometer. CO2 removed by the animal's lungs was determined by the expired gas CO2 content in a Dou glas bag, We made stepwise, 20% reductions in ventilator support hourl y. We first reduced the tidal volume to achieve a peak inspiratory pre ssure of <30 cm H2O, and then we reduced the respiratory rate while ma intaining normocapnia. Pao(2) was maintained by adjusting the FIO2, an d the level of positive end-expiratory pressure. Measurements and Main Results: Mean blood flow through the arteriovenous shunt ranged from 1154 +/- 82 mL/min (25% cardiac output) to 1277 +/- 38 mL/min (29% car diac output) over the 6-hr study period. The pressure gradient across the gas exchanger was always <10 mm Hg. Maximum arteriovenous CO2 remo val was 102.0 +/- 9.5 mL/min (96% of total CO2 production), allowing m inute ventilation to be reduced from 10.3 +/- 1.4 L/min (baseline) to 0.5 +/- 0.0 L/min at 6 hrs of arteriovenous CO2 removal while maintain ing normocapnia. Similarly, peak inspiratory pressure decreased from 4 0.8 +/- 2.1 to 19.7 +/- 7.5 cm H2O. Pao(2) was maintained at >100 torr (>13.3 kPa) at maximally reduced ventilator support. Mean arterial pr essure and cardiac output did not change significantly as a result of arteriovenous shunting. Conclusions: Extracorporeal CO2 removal using a low-resistance gas exchanger in a simple arteriovenous shunt allows significant reduction in minute ventilation and peak inspiratory press ure without hypercapnia or the complex circuitry and monitoring requir ed for conventional extracorporeal membrane oxygenation. Arteriovenous CO2 removal can be applied as an easy and cost-effective treatment to minimize ventilator-induced barotrauma and volutrauma during severe r espiratory failure.