|
|
October 16-20, 2010 San Diego, CA Home Abstract Archive Search Abstracts 2010 Session Grid 2010 Meeting Website Policy Statements ASA Website Feedback 
Visit Anesthesiology.org |
| Previous Abstract | Next Abstract |
| Printable Version |
|
A1469
October 21, 2009 9:00 AM - 11:00 AM Room Area F |
| Dynamical Properties of a New Transcutaneous Carbon Dioxide Sensor |
|
* Volker Hartwich, C.R.N.A., Peter M. Schumacher, M.Sc., Ph.D., Antonello Caruso, M.Eng., Martin Luginbuehl, M.D., Ph.D. Department of Anesthesiology, University Hospital Bern, Switzerland |
|
Background
Sedative or analgesic drugs are often used in medical interventions in spontaneously breathing patients, the potentially most dangerous side effect of those drugs is respiratory depression[1]. Transcutaneous CO2 (tcCO2) monitoring is fast and reliable to detect the onset of respiratory depression, compared to measurement of oxygen saturation (SpO2) while administering supplemental oxygen[2]. We compared the latest Sentec digital V-Sign2 sensor and its predecessor V-Sign and endtidal CO2 measurements in a dynamic situation. Methods After IRB approval and informed consent 10 healthy volunteers were enrolled for a CO2 rebreathing study. Sentec V-Sign and V-Sign2 digital sensors (MPB software v04.04.04 and v05.00.00) with combined SpO2 and tcCO2 measurement were attached to the volunteers earlobes (SenTec AG, Switzerland). Subjects started to breath through a tight fitting mask with attached spirometry and side-stream capnography. After stabilization a bag filled with 7 L oxygen was attached to the open spirometry tube and rebreathing started until the subject felt uncomfortable. The bag was then removed and breathing was monitored until baseline values were reached again. For sensor response modeling high resolution recordings of endtidal CO2 partial pressure (PetCO2) and PtcCO2 from the two sensors were fed into a previously identified model [3]. A nonlinear regression approach was used for parameter estimation. The step answer of the model was calculated to define the time to reach 90% of end value (T90) and the steepest slope (SL) and compared to the apneic situation (assumed rise of 6 mmHg/min of PaCO2 following a fast 7 mmHg arterial mixed-venous equalization [4]). Results The average rebreathing period was 4.6 ±1.3 minutes, max. PetCO2 was 59 ±8 mmHg, max. PtcCO2 was 50 ±7 and 53 ±7 for old and new sensor respectively. A two compartment model of the form dP1/dt = -k12*P1 + k21*V2/V1*(P2-PSS) with an additional lag time and processing related averaging function fitted the data best (P1=pCO2 in sensor, P2=PetCO2, PSS=steady state offset). After visual inspection of the data the lag and averaging times of the sensor were fixed at 26 and 25 s and at 15 and 16 s for the old and new sensor, respectively. The rate constants k12 and k21*V2/V1were 0.67 [0.56 0.76] and 1.27 [1.16 1.48] min-1 and PSS was 0.45 [-0.69 1.73] and 0.66 [-0.86 1.84] mmHg (median [quartiles]) for old and new sensor. The median fitting errors expressed as MDPE (median prediction error, i.e. bias) and MDAPE (median absolute prediction error, i.e. inaccuracy) for individual and population fit were all smaller than 5%. Simulated step answer (PetCO2 increase from 40 to 60 mmHg) resulted in a T90 of 3.8 / 2.1 minutes and a SL of 11.8 / 22.0 mmHg/min for old and new sensor, respectively. The sensors shows a 4 mmHg increase within 58 and 36 s after simulated apnea, old and new respectively. The old sensor was 4 years old and due to aging considerably slower as shown previously [3]. Conclusion The new V-Sign2 sensor was significantly faster than the older V-sign sensor. A change rate of up to 22 mmHg/min allows for accurate CO2 monitoring even in case of fast changing respiratory conditions such as apnea. References [1] Anesthesiology. 2006 Feb;104(2):228-34. [2] Respir Care 2003; 48:611-20 [3] Anesthesiology 2006; 105: A477 [4] J Clin Anesth. 1988;1(2):96-103. From Proceedings of the 2009 Annual Meeting of the American Society Anesthesiologists. |