Mr Tolin

I am not sure whether is appropriate to use norms of blood gas analysis as norms for capnography (PaCO2 and ETCO2 are in reality different variables although they are closely related). Validity studies about capnography occasionally suggest that it underestimates the value of PaCO2 (Hinkelbein, 2008; Liu et al, 1992; Stein, 2006) though in non-healthy populations. According to the book by Hockenberry and Wilson (2015, p 1140) normal values of ETCO2 are 30-43 mmHg which is slightly lower than arterial PaCO2 (35-45mmHg). I suppose that the model (e.g. BIOPAC) used may additionally affect the normal values. Based on this your findings may not be wrong (however I don't know if the same happens by using the BIOPAC). I hope that this answer in parallel with other answers that will follow will provide some directions. Good luck with your work...

Hockenberry M.J. and Wilson D. (2015). Wong’s nursing care of infants and children. Canada: Elsevier Mosby

Hinkelbein, J., Floss, F., Denz, C., Krieter, H. (2008). Accuracy and precision of three different methods to determine Pco2 (Paco2 vs. Petco2 vs. Ptcco2) during interhospital ground transport of critically ill and ventilated adults. Journal of Trauma 65, pp. 10-18.

Liu S.Y., Lee T.S. Bongard F. (1992) Accuracy of capnography in nonitubated surgical patients. Chest 102(5), pp. 1512-1515

Stein N., Matz H., Scheeweib A., Eckmann C., Roth-Isigkeit A., Huppe M., Gehring H. (2006). An Evaluation of a Transcutaneous and an End-Tidal Capnometer for Noninvasive Monitoring of Spontaneously Breathing Patients. Respiratory Care 51(10), pp. 1162-1166

Thanks for your reply. It's weird that I see so many sources use the 35-45 mmHg guideline for ETCO2, even though ETCO2 runs lower than PaCO2. I notice that the Hockenberry resource is for children though; should the range in adults be the same?

I think that although their book is about infants and children, their comment was about ETCO2 in general. Unfortunately, their comment is not referenced for detecting the original study. However, there are booklets and presentations discussing about these normal values (though unpublished and not appropriate to be used as references). I was not able to find a normative study in the past for both ETCO2 and PtcCO2 (transcutaneous blood gas assessment) and it is strange that such studies are missing in healthy people (according to my knowledge). I would suggest if you finally do not detect such a study to contact company of BIOPAC to ask them about the means and SDs in ETCO2 during the testing of the equipment. If they do not provide you with satisfactory answers, then you could proceed with publication of your results by supporting indirectly the findings based on a) the general underestimation of PaCO2 by capnography (although in reality does not assess the PaCO2), b) the different normal values provided in the book I mentioned and c) being cautious about the interpretation of the results.

Please additionally have a look at the following capnography reference handbook which additionally provide the same normal values (without being specific for children):

http://www.mysupplies.philips.com/Documents/CapnographyReferenceHandbook_OEM1220A.PDF

Kind regards

Zach

Hello David,

We primarily worry about engineering more than science.  We want to make sure, for instance, that if you succeed in getting an appropriate gas sample into the CO2100C, the system will provide you with accurate information about the concentration of carbon dioxide in that sample.  It is not necessary to obtain normative physiological data for that goal.  

I would suggest that your best path forward would be to contact colleagues who have used our equipment to perform similar measures.  A Google search on "CO2100C ETCO2" will find you several papers describing research using our equipment.  Most do find EtCO2 levels that are not that far below PACO2 levels.  

Although you can find papers indicating EtCO2 should be significantly lower (e.g., http://rc.rcjournal.com/content/respcare/58/7/1152.full.pdf -- note that this paper is not one I found in the above-mentioned search; this group did not use our equipment), you can also find papers saying it should be close (e.g., https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1062759/pdf/jcinvest00343-0045.pdf)

Researchers using our equipment do seem generally to find higher values than you are reporting:

http://www.pnas.org/content/109/21/8270.full.pdf (Figure 3B indicates subjects breathing room air normally had average EtCO2 of 5.9%)

http://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0163071&type=printable (Figure 2B and Table 1 find EtCO2 levels around 40 mm Hg when not experimentally manipulated)

http://www.sciencedirect.com/science/article/pii/S1053811914008568 (Figure 1A shows EtCO2 values generally between 35 and 40 mm Hg)

https://pdfs.semanticscholar.org/e029/a805c95ef846307cc1178581bc15e5b1b361.pdf (Table 1 indicates baseline EtCO2 of 40 mm Hg)

There are at least a few others, but those will get you started.

I hope you find this useful.

Mickey

Thank you Mickey-- very glad to get your response! So if other papers using the BIOPAC are getting significantly higher ETCO2 values than we are, it makes me wonder whether (a) we just had some oddly hypocapnic people in our sample or (b) something is off with the equipment.  We'll collect a bit more data on healthy controls for now, and see what we come up with.

Hi David,  I used to do quite a bit of research with end-tidal capnography. Part of the problem may possibly have to do with your quantification procedure: not all breaths show an end-tidal plateau. If you are merely automatically averaging all the breaths, you are bound to get underestimations, since artefactual values (i.e. more or less spikes--but not plateaus--in the signal will be of lower magnitude. Try to eliminate those, if that is what you are doing, and see wqhat happens. I'd be curious for a reply.

Also the range of traditional values were taken from arterial levels, and i think there is good reason to believe that even the plateau levels of end-expiratory CO2 don't reach true expiratory levels (check it out yourself by breathing slower (and somewhat deeper) with the sensor on (but make sure you don't start hyperventilating (your minute ventilation volume increases from normal resting pattern of breathing).

Dear David

As Paul suggests, the likeliest reason for your values is that the healthy normal subjects are hyperventilating (e.g. due to situation anxiety, pregnancy, cold air).

Another reason might be anatomical and apparatus dead space: with normal quiet respiration you are unlikely to be getting a true end-tidal (alveolar) sample from the lips. Even with endotracheal anaesthesia, there is usually about a 2 mmHg difference between ETCO2 and Arterial PCO2.

Also, I note from the biopac info that the CO2 module is fitted with a variable speed sampling pump; for accurate real time CO2 readings this should be set at the fastest recommended sampling rate (150-200 mL/min for an adult, 50 for an infant).

Finally, how long is your sampling tube and are you in an air-conditioned lab? The length of the tube usually doesn't matter - it just delays the measurement for each breath, but if the environmental temperature is low, condensation of water in the sampling line can screw up your measurements.

Ideally you should be taking your values from a calibrated paper trace, using an analog signal (sorry, showing my age!).

Good luck

Paul and Christopher, thank you for your responses!

Paul, I hadn't thought about removing breaths without an end-tidal plateau.  I'll look into that.

Christopher, we switched out from the mask to a nasal cannula, 3 feet or so of tubing.  We've ordered calibration gas so we can get a better idea of what's going on. We'll check the pump speed as well.

Best,

David

David

All the best. Actually, I feel a bit of a prune. I forgot it 's deepest winter in the Northern Hemisphere and I was talking about Aircon! You're probably knee-deep in snow!

Cheers

Yes, quite cold here.  We're looking at room temperature but my guess is that it's neither too warm nor too cold to explain the findings.

I did consider the possibility that our sample was hyperventilating-- but we cross-tested some of them on the Freespira device, and the BIOPAC numbers were lower so I think it has something to do with our specific equipment.

Hi David,

Agree with comments above, your values for baseline EtCO2 do seem a little low. I would like to point you in the direction of work by some of my colleagues' on normative physiological measurement data in a large cohort (n=119):

Article The Leicester cerebral haemodynamics database: Normative val...

This demonstrated median EtCO2 values of 39.1(IQR 36.9–40.8) and mean 38.6 (SD 3.9). This is certainly inkeeping with studies using a basic nasal cannulae setup/face mask for TCD measurements.

However, be wary of the variation that exists between cannulae and facemasks as shown by some of my previous work:

Article PaCO2 measurement in cerebral haemodynamics: Face mask or na...

Hope that helps,

Best wishes

Jatinder

Many thanks!

I am sorry, I have not been studied your request.