Dear Dr Rahman and Colleagues,

I have read with great interest your recent report of the toxicity, oxidative stress, and inflammatory response of electronic cigarettes vapors and I have a few queries and observations.

1. The article refers to the use of a Vision Spinner e-cigarette battery but does not mention the voltage that the battery (and hence power of the heating coil) that the used during the study. The Vision Spinner I have found online can operate from 3.3V to 4.8V which when used with the 2.2 Ohm coil you used would give coil power from 4.95 Watts to 10.5 Watts. Can you clarify what voltage was used for these studies.

2. In your tests of OX/ROS production in table 2 of your study I notice that the higher viscosity e-juices ("Glycerine" and "Propylene Glycol:Glycerine (50:50)") show very high variability between individual tests. For the glycerine test the run with the highest H₂O₂ equivalent is 248% of the result for the lowest test; for the 50:50 mix the highest is 281% of the lowest; for the propylene glycol the highest is only 107% of the lowest. These data do not show a high level of reproducibility of the measurements of OX/ROS.

I am concerned that this variability of OX/ROS measurement may be due to the device entering a "dry puff" operating regime that does not accurately reflect real-world usage. Essentially a "dry puff" is where the heating coil overheats due to insufficient wicking of e-juice or poor contact between wick and coil. This is exacerbated by high power application to the heating coil and by viscous e-juices such as those that have significant concentrations of glycerol. Intuitively one would expect greater OX/ROS production in this operating regime. In real-world use the dry-burn state is readily detected by its unpleasant taste which informs the user that something is amiss with the device.

The recent articles on formaldehyde production in ENDS [1] have been criticised [2] for running the devices in the "dry puff" regime. I wonder whether you were aware of this and if/how you controlled for it.

3. Your "dripping" experiments. As I understand it you dripped 2 drops of eliquid on the coil and then passed 20 puffs through the DCFH solution. If so, it would seem inevitable that the eliquid is going to run out during the test resulting in the coil overheating. I don't think that your tests on 'dripping' in represent the reality of their usage.

When dripping, a purpose-designed atomiser is used with a well into which is dripped 10-20 drops of eliquid. The heating coil has wicks dangling into the well. Examples can be found if you search Google Images for "RDA build" [3]. Most importantly, users regularly refill the well to prevent "dry puffs".

I would be interested to hear your thoughts on the above points.

Kind regards... Mark Dobson (PhD)

[I have no competing interests]

[1] Jensen et al. (2014) NEJM. http://www.nejm.org/doi/f...
[2] Dr. Farsalinos's blog. http://www.ecigarette-res...
[3] http://images.google.com/...