1995 OPEN FORUM Abstracts
DOES SAMPLING RATE EFFECT THE RELATIVE AGREEMENT OF PetCO2 MEASURED BY RAMAN AND MASS SPECTROMETRY?
J Totaro, JM Graybeal, CRTT, GB Russell, MD. Dpt of Anesth, PSU College of Medicine, Hershey, Pa. 17033
INTRODUCTION: Raman Spectrometry (RS) and Mass Spectrometry (MS) are two common methods used to measure respiratory gases, including end-tidal CO2 (PetCO2). The bias between these techniques is effected by respiratory rate (RR) (Respiratory Care 94:190-194). We determined to test the following hypothesis; the bias between PetCO2 measured by RS and MS will be minimized by altering the sampling rate (Samp) of the RS, which is 220 ml/min normally.
Methods: Using the method previously described, a bench top lung model, with CO2 added, simulated respiration. An airway connector with multiple sampling ports allowed simultaneous sampling by calibrated RS and MS. PetCO2 was determined in triplicate at each of these Samp MS (60 ml/min) and RS (220, 156, 131, 102 ml/min), RR (0, 10, 18, 28, 35, 40 bpm) and various CO2 levels. Data was analysized by linear regression, and ANOVA with Scheffe post hoc analysis. A p valueɘ.05 determined significance. The bias and limits of agreement were calculated for each Samp.
Results: 60 pairs of PetCO2 were collected with a positive correlation between the RS and MS (r = 0.98, slope = 1.07). (All values are mean (SD)). PetCO2 ranged from 26.6 to 54.7 mmHg (40.5 (11.1) for RS and 23.2 to 53.7 mmHg (36.7 (10.2)) for MS. The bias between RS and MS was 3.83 (2.11) mmHg. The bias was effected by RR, with significant differences between RR's 10,28 (p=0.0095), 10,35 (p=0.001), and 10,40 (p=0.0003). (Figure 1) No significant differences were found between the different Samp. (Figure 2)
Conclusions: Altering RS Samp does not effect the RS-MS bias. The cause of this bias remains unclear, and may be due to the different algorithms determining PetCO2 between the RS and MS.
SEE ORIGINAL GRAPH