Value and accuracy of dual oximetry during pulmonary resections

Abstract During thoracic surgery, one-lung ventilation (1 LV) is often required. The purpose of this prospective study was to examine the usefulness and accuracy of dual-oximetry during 1 LV. Prior to the induction of anesthesia, 30 patients had a radial artery and a fiberoptic pulmonary artery catheter (15 Edwards, 15 Spectramed by randomization) inserted. Arterial O 2 saturation (SPO 2 ) was monitored by pulse oximetry, and mixed venous O 2 saturation (SvOin2) by oximetry (Edwards or Spectramed). Arterial and mixed venous blood gases were obtained and immediately analyzed by an OSM3-Hemoximeter. Measurements, including hemodynamics and blood gases, were obtained before induction, during two-lung ventilation (2LV) in the supine and lateral decubitus positions, during 1 LV, and following extubation. The change from 2LV to 1 LV was associated with significant increases in cardiac index (CI) and oxygen delivery index (DO 2 I, whereas PaO 2 and arterial and mixed venous oxygen saturation decreased. The ratio of oxygen consumption to delivery remained stable. Continuous oximetry when compared with in vitro measurements yielded a correlation coefficient for arterial oxygen saturation of r = 0.794 ( P r = 0.874 ( P /≤ 0.001) and −1.3% ± 2.8% for the two-wavelength Edwards catheter; and, r = 0.862 ( P ≤ 0.001) and −0.1% ± 3.2% for the two-wavelength Spectramed catheter. These findings demonstrate that dual-oximetry is an on-line, reliable method to measure SpO 2 and SvO 2 . SPO 2 2 ≤ 70 mm Hg). SvO 2 is determined primarily by oxygenation ( r = 0.005; P ≤ 0.05) rather than by CI (r = 0.001, ns). Since DO 2 I increased during 1 LV to maintain the oxygen supply and demand balance, SvO 2 monitoring might be useful as an early indicator in identifying high-risk patients with compromised DO 2 I resulting from decreased CI.