Growing knowledge in avian medicine has resulted in more sophisticated and challenging surgeries with longer anesthetic time. Simultaneously the requirement and need for better anesthetic monitoring has increased. Currently anesthetic monitoring of critical avian patients includes the measurement of standard pulse oximetry (SpO2) for monitoring arterial hemoglobin oxygen saturation. Especially in longer anesthesias, postoperative recovery and critical care monitoring of ventilation may be as important as the assessment of oxygenation. Blood gas analysis, the gold standard of oxygenation assessment, is generally not available due to small size and blood volume in avian species. Newer investigations in grey parrots (Psittacus erithacus) have shown that end-tidal pressure of carbon dioxide (EtCO2) would be a reliable indirect measurement of arterial carbon dioxide tension (PaCO2).2 In addition, recent developments in transcutaneous carbon dioxide (tcPCO2) monitoring have shown promising results in different mammalian species1,7 and is a well-accepted monitoring tool in human intensive care.4 The objective of the current study was to determine whether tcPCO2 measurements will reflect actual CO2 status in anesthetized avian species and to test a transcutaneous sensor for the first time in an avian species.
Thirty-five healthy Lohmanns selected leghorn chickens (Gallus gallus) were anesthetized with 4% isoflurane (Attane™, Provet AG, 3421 Lyssach, Switzerland) in oxygen delivered by face mask and intubated by endotracheal tubes. Anesthesia was maintained by administration of isoflurane (end-tidal concentration of isoflurane: 1.1–1.3%). Animals were instrumented for end-tidal (Cardiacap/5, Datex-Ohmeda, Helsinki, Finland) and transcutaneous carbon dioxide tension (V-Sign™ Sensor, Sentec AG, Basel, Switzerland) monitoring. Arterial blood for PaCO2 analysis was collected with preheparinized syringes from arterial puncture of the A. ulnaris or A. tibialis cranialis, gently mixed and tested without delay with a portable clinical analyser (i-STAT, Heska AG, Fribourg, Switzerland).
The analytic performance of the transcutaneous biosensor (tcPCO2) was compared with EtCO2 measurement, and arterial PaCO2. Comparison of tcPCO2, EtCO2 and PaCO2 results were performed according to standard analytic techniques, based on Deming’s regression and Bland-Altman bias representation3,5 using a personal computer-based statistics software (GraphPad Prism, version 4.00, GraphPad Software, San Diego, CA, USA). Significance level was set at p[CLKB1]=0.05.
The transcutaneous CO2 sensor produced reliable results in 87.5% (28/32) of investigated animals, while arterial blood collection was possible in 85.7% (30/35) and end-tidal CO2 measurement in 100% (35/35) of investigated animals. Mean and standard deviation (SD) of tcPCO2 (25.08 mm Hg±14.17) were similar to PaCO2 (24.06 mm Hg±6.91), but lower than EtCO2 (28.91 mm Hg±6.36). Results from tcPCO2 and PaCO2 (r=0.45, p=0.02), tcPCO2 and EtCO2 (r=0.64, p<0.001), and EtCO2 and PaCO2 (r=0.56, p<0.001) correlated with each other. Deming’s regression slopes were close to 1.00 and intercepts close to zero for all three parameters. Slopes were significantly different from zero in all three analyses. Overall accuracy was acceptable. Bias of EtCO2 and tcPCO2 comparison was smaller than in tcPCO2 and PaCO2 and in EtCO2 and PaCO2 comparisons. Limit of agreement was fairly wide in all three CO2 measurement comparisons.
In summary, transcutaneous CO2 measurement is technically easy to perform, non-invasive, and does not increase airway resistance. The tcPCO2 sensor was reliably tested in humans6 and sheep.7 The current study expands its application to avian species as an indirect measurement of adequate ventilation. The wider range of tcPCO2 in comparison to PaCO2 and EtCO2 measurements indicates that single results have to be interpreted carefully. In our experience, attention has to be paid to carefully depluming the avian skin to improve sensor contact area and to improve adequacy of results. Similar to a previous study,2 EtCO2 overestimated PaCO2 by approximately 5 mm Hg.
Although Deming’s regression analysis of the different comparisons of CO2 measurements resulted in significant correlation, results have to be interpreted with caution, because Bland-Altman analysis revealed, besides an acceptable bias, a fairly wide range of differences within the 95% limits of agreement. Since these differences are clinically important, we conclude that the three methods may be used interchangeably only with caution in avian species. Nevertheless, tcPCO2 measurements offer additional information for monitoring ventilation when EtCO2 or PaCO2 measurements are not available or feasible. Further research in critical patients is needed to test its clinical reliability.
The authors thank Sentec AG, Switzerland for providing the V-Sign™ oxygen saturation and transcutaneous carbon dioxide tension device and Sandra Mosimann, Sarah Scharmer, and Ramiro Valero for all their help during the study.
1. Abitbol, M.M., A.G. Monheit, and M.L. Stone. 1986. Arterial PO2, PCO2, and pH versus transcutaneous PO2 and PCO2 and tissue pH in the fetal dog. Am. J. Obstet. Gynecol. 155: 437–443.
2. Edling, T.M., L.A. Degernes, K. Flammer, and W.A. Horne. 2001. Capnographic monitoring of anesthetized African grey parrots receiving intermittent positive pressure ventilation. J. Am. Vet. Med. Assoc. 219: 1714–1718.
3. Jones, R.G., and R.B. Payne. 1997. In: Clinical Investigation and Statistics in Laboratory Medicine, ACB Venture Publications, London, UK. Pp. 27–65.
4. Kocher, S., R. Rohling, and A. Tschupp. 2004. Performance of a digital PCO2/SPO2 ear sensor. J. Clin. Monit. Comput. 18: 75–79.
5. Marquis, P. 1999. [Comparisons of analytical methods]. Ann. Biol Biol.Clin. (Paris). 57: 737–738.
6. Rohling, R., and P. Biro. 1999. Clinical investigation of a new combined pulse oximetry and carbon dioxide tension sensor in adult anaesthesia. J. Clin. Monit. Comput. 15: 23–27.
7. Vogt, R., R. Rohling, and S. Kästner. 2004. A new combined transcutaneous carbon dioxide tension and pulse oximetry sensor: a preliminary evaluation in adult sheep. Proc. AVA Autumn Meeting, Vienna. 95.