Our comments in reply to the arguments presented in the CrossTalk articles by Teppema & Smith (2013) and Wilson & Day (2013) are confined to experiments on humans, and in their CrossTalk article Teppema & Smith (2013) cite two studies that found evidence supporting a hyperadditive interaction. First, carotid body denervation or removal in man has consequences for the central chemoreception of CO2. We agree that such denervation removes the tonic drive from the carotid bodies (Dahan et al. 2007), evident from their involvement in sympathetic tone. However, if this drive is indeed tonic, then it does not change with CO2 and so does not function as a hyperadditive factor during hypercapnia. With respect to species differences, it is interesting to note that carotid body denervation in dogs has little effect on the normoxic ventilatory response to CO2 (Mitchell, 1965).
The second human study cited by Teppema & Smith (2013) involved acid–base changes induced with acetazolamide and intravenous bicarbonate (Teppema et al. 2010). While arterial tensions and [H+] were determined in these experiments, the stimulus to the central chemoreceptors is unknown, and the effects of acetazolamide on the cerebrovascular sensitivity to CO2 (Fan et al. 2012), and the central and peripheral chemosensitivity to CO2 (Vovk et al. 2000) are confounding factors that make interpretation problematic.
There are two common elements in the arguments presented by Wilson & Day (2013) and Teppema and Smith (2013): the recognition that species differences may play a role, and that the various experimental models and scenarios devised to test the interaction between central and peripheral chemoreflexes are complex and subject to limitations. We agree and suggest that a simple test is needed that can be applied across species so that comparisons may be made.
The most appropriate experimental preparations therefore need to be identified and employed to move the field forward. For example, vagotomized and decerebrate preparations may have a place in answering specific questions, but it is unlikely that the responses to CO2 and O2 in this type of a preparation would be comparable to a spontaneously breathing intact animal. As we explained in our CrossTalk article (Duffin & Mateika, 2013), the hypoxic, isoxic ventilatory response to hypercapnia can determine the type of interaction between the peripheral and central chemoreflexes. We suggest that this response is the appropriate pan-species test because it is applied to the intact animal and yields the ventilatory regulator characteristics under physiological circumstances.