51 Monitoring: Equipment and Computers ing 3 minutes af ter carotid artery cross clamp were also comparable: rSO- 2 was 74 (65-81) in Gabs and 75 (59-90) in Gdef (p=0.70), the calculated drop ǻrSO 2 min was 6 (2-13) in Gabs and 2 (0-6) in Gdef (p=0.15). rSO 2 avg and ǻ rSO 2 avg did not dif fer bet ween groups either. Similarly, there was no dif- ference bet ween groups on contralateral side. No influence of non-operated (contralateral) carotid ar ter y stenosis on rSO 2 was proved. Conclusion(s): Neither absolute rSO 2 values nor rSO 2 drop af ter carotid ar- tery cross clamp dif fered bet ween patients with and without neurological deficit. rSO2 was not reliable in prediction of cerebral ischemia during CEA. 3AP6-1 Calibrated vein diameter measurement using vein illumination technology Joke D.B., Van Poucke S., Frank J. ZOL Genk, Dept of Anaesthesiology & Intensive Care, Genk, Belgium Background and Goal of Study: Of all invasive medical procedures, veni- puncture is the most common but at tempts can fail. Dif ficult draws include those from babies, dark-skinned and obese patients. Hemoglobin in the blood absorbs infrared light. When a vein illumination device such as the AccuVein AV300 ® ( Accuvein Inc Huntington, NY) is held above the skin, veins appear noticeably dif ferent than the surrounding tissue aiding in vein location. The goal of the study was to quantitatively assess the vein diameter with and with- out the AccuVein technology. Materials and Methods: Digital images of 77 human forearms (ventral side) plus calibration char t were taken using a digital camera, with and without the Accuvein device. All images were planimetrically and colorimetrically calibrat- ed as previously published 1 . During calibration, using uniform illumination and a reference chart as part of the image of interest. The calibration provides a means of transforming the acquired images, to a standard, well-defined co- lour space i.e. SRGB. Vein diameters were measured using ImageJ, (NIH). Comparison of the t wo methods were statistically assessed using Passing and Blalock regression and illustrated by a Bland and Altman plot. 2,3 Results and Discussion: Passing and Blalock regression reveiled a regres- sion equation of: y=-0.09111+1.2593x with a 95% CI of intercept A bet ween -0.2641 and 0.0120 and a 95% CI intercept B bet ween 0.9000 and 1.7727. A Cusum test for lineairit y demonstrated no significant deviation from lineairit y (P>0.05). Conclusion(s): Although perceived as a visual aid during puncture, no signifi- cant dif ference of vein diameter was measured. Future research could inves- tigate the influence of dif ferent skin colours on the visibilit y of veins and the success rate of punctures using the AccuVein device. References: 1. Automatic colorimetric calibration of human wounds. Sven Van Poucke, Yves Vander Haeghen, Kris Vissers , Theo Meer t and Philippe Jorens. BMC Medical Imaging 2010, 10:7. 2. Bland JM, Altman DG (1986) Statistical method for assessing agreement bet ween t wo methods of clinical measurement. The Lancet i:307-310. 3. Passing H, Bablok W (1983) A new biometrical procedure for testing the equalit y of measurements from t wo dif ferent analy tical methods. Application of linear regression procedures for method comparison studies in Clinical Chemistr y, Par t I. J. Clin. Chem. Clin. Biochem. 21:709-720. 3AP6-2 Inspired O 2 concentrations when O 2 /N 2 O and O 2 /N 2 fresh gas mix tures are used at the ox ygen ratio controller limits of the Zeus ® anesthesia machine Schollaer t C., De Cooman S., Struys M., De Wolf A., Hendrick x J., NAVAt group NAVAt group - OLV Hospital, Dept of Anaesthesiology & Intensive Care, Aalst, Belgium Background and Goal: The S-ORC hypoxic guard of the Zeus ® (Dräger, Lübeck, Germany) is more stringent than required by existing standards. We measured the inspired O 2 fraction (F I O 2 , in %) that resulted from using O 2 /N 2 O or O 2 /N 2 (= air) mix tures in conventional mode with the delivered FO 2 (F D O 2 , in %) set at the S-ORC limit across a wide fresh gas flow FGF range. Materials and Methods: Af ter IRB approval, 16 ASA I-II patients received sevoflurane in either O 2 /N 2 O (n=8) or O 2 /N 2 (n=8). Af ter using an 8L/min FGF with F D O 2 = 25% for 10 min, the following S-ORC limits were tested for 4 min each: 0.3[85]; 0.4[65]; 0.5[50]; 0.7[36]; 0.85 [30]; 1.0[25]; 1.25[25]; 1.5 [25]; 2 [25]; 3[25]; 5[25];8 [25]; first number = total FGF (L/min), second number in bet ween brackets =F D O 2 (%). Bet ween combinations, 8L/min FGF with 25% O 2 was used for 4 min. The sequence was studied once with O 2 / N 2 , but t wice in the same patient with O 2 /N 2 O (to examine the ef fect of decreasing N 2 O uptake over time), resulting in 3 groups: early N 2 O, late N 2 O, and N 2 group. If F I O 2 remained t 21% in all the patients for a par ticular ORC limit (=FGF [F D O 2 ] pair), we repor t the F I O 2 after 4 min for that pair. If F I O 2 became < 21% in even a single patient in a par ticular FGF[F D O 2 ] group, we report the % of patients with F I O 2 < 21%, and the time (min) af ter which this F I O 2 % was reached in this subset of patients. Data are presented as average ± standard deviation. Results and Discussion: Early N 2 O Late N 2 O N 2 ORC limit: FGF[F D O 2 ] Incidence of F 1 O 2 <21% within 4’ ｨW WR F 1 O 2 <21% Incidence of F 1 O 2 <21% within 4’ ｨW WR F 1 O 2 <21% Incidence of F 1 O 2 <21% within 4’ ｨW WR F 1 O 2 <21% L/min[%] % of pts min % of pts min % of pts min 0.3 [85] 0 0 0 0.4 [65] 0 0 0 0.5 [50] 0 0 0 0.7 [36] 0 0 17 3.3 * 0.85 [30] 0 38 3.4 (0.5) 25 2.3 (1.4) 1 [25] 88 2.2 (0.7 100 1.6 (0.5) 100 2.1 (0.8) 1.25 [25] 100 2.1 (0.9) 100 1.4 (0.5) 100 1.6 (0.3) 1.5 [25] 100 2.2 (0.8) 100 1.7 (0.8) 100 2 (0.7) 2 [25] 63 2.7 (0.5) 75 2.3 (0.6) 38 1.5 (0.3) 3 [25] 0 13 3.2 * 0 5 [25] 0 0 0 8 [25] 0 0 0 [Figure 1. Legend: * = 1 patient only ] The overall F I O 2 -FGF relationship in the 3 groups was similar (Figure 1): af ter using a par ticular FGF for 4 min, F I O 2 (1) was highest with the lowest FGF (0.3 L/min) [not presented]; (2) progressively decreased when a higher FGF was used [not presented]; (3) became < 21 % in some patients with a 0.7 to 3.0 L/ min FGF and < 21 % in all patients with a 1 and 1.25 L/min FGF; and (4) gradu- ally approached F D O 2 with FGF > 1.25 L/min. In the late N 2 O and N 2 groups, F I O 2 became < 21% faster, and over a wider FGF range. Dif ferences bet ween the late N 2 O and N 2 groups were small. Conclusion: Even a hypoxic guard more stringent than required cannot pre- vent the formation of hypoxic inspired mix tures. 3AP6-3 Preliminar y development and evaluation of a suppor t system for care of mechanically ventilated patients (SCMVP) Andoh T., Maruyama K., Miyaji T., Hasegawa T., Eguchi K. Mizonokuchi Hospital Teik yo University School of Medicine, Dept of Anaesthesiology, Kawasaki, Japan Background and Goal of Study: To ease identification of the site of a respira- tor y circuit leak, we have developed a novel alarm system, the suppor t system for care of mechanically ventilated patients (SCMVP), which provides graphi- cal information on the site of the leak. We preliminarily evaluated usefulness of the SCMVP protot ype in a mechanically ventilated manikin. Materials and Methods: The SCMVP prototype consists of several compo- nents, including a personal computer and sensors for gas flow, CO 2 , pressure, and misconnection of respiratory circuits. Integrated information from these sensors is displayed on a panel to indicate the site of leakage. Fif teen reg- istered nurses were recruited to evaluate the respiratory circuit with SCMVP (SCMVP system) and without SCMVP (Conventional system). A manikin was mechanically ventilated in a volume control mode. Its trachea was intubated and lungs were insuf flated with CO 2 . We developed four leakage scenarios based on near-miss/adverse event repor ts 1 . The scenarios included a leak at tracheal tube cuf f, unplanned ex tubation. We randomly assigned t wo sce- narios to the Conventional system and the other t wo scenarios to the SCMVP for each par ticipant. Time to identif y the site of leak was measured and com- pared bet ween t wo systems using the Wilcoxon signed-rank test. Results and Discussion: The SCMVP system showed significantly shor ter time for troubleshooting in one of four scenarios, a leak at the Y-connector, and shor ter accumulated time for troubleshooting compared with the Conventional system. SCMVP resulted significantly more frequent incidences of troubleshooting within 30 sec (43.3% (13/30) with SCMVP vs. 14.4% (4/30) with Conventional). It also showed significantly less frequent incidence of trou- bleshooting of >180 sec (6.7% for SCMVP vs. 30% for Conventional).