Comments are closed for the article. So here are my opinions.
First off, I do not fault a journalist for trying to communicate a known important problem in vaping. This problem is potentially very serious, and people do need to know about DA and AP. If I did not believe that I would not have been part of the team that first published this a year ago. Nor do I at all fault Dr. Reid of Marquette for taking an honest crack at testing for these compounds. I applaud his efforts, and I hope he can bring another validated method online in this big effort to make vaping as safe as possible. But there are some very profound problems with many details of this article out of the Milwaukee Journal Sentinel by Raquel Rutledge.
From the article:
"
Marquette and Enthalpy said their analyses passed their quality control measures and that they stand by their results.
The discrepancy highlights differences in testing methodology and desired levels of detection. Marquette's approach used a more sensitive technique that scientists say can detect lower levels, down to parts per billion. The Journal Sentinel consulted with three independent experts who said the university's method was superior.
The problem is, it is also more time-consuming and costly."
No, this is NOT the problem, and the author has been informed of this, because Enthalpy has contacted her and tried to explain the science. A lower LOD (Limit of detection) or LOQ (limit of quantification) is certainly preferred, but ONLY if the method assay is validated against other methods, and the lab has a lot of experience doing this particular test. Diketones are notoriously difficult to get accurate levels of, and it can take a
lot of time and effort to get trustworthy data, especially when there are other flavor compounds present, as in an e-liquid. WAY more time and cost than is implied that Marquette went through for their 5 samples, which are the only samples that they have tested for DA and AP, as in
ever. The difference between an LOD of 1 ppm or 0.5 ppm is NOT significant, and certainly does not define a "superior" method, especially if the following information is taken into account.
The Marquette lab (Dr. Scott Reid) used GC-MS with a derivative method (converts DA and AP to specific nitrogen compounds that are then analyzed). This is fine, but they had NEVER done this assay before, and it is not a validated method. Meaning they never did any other known valid method for checking their numbers. Nor did they use an internal standard to measure against which is common practice for qualitative GC-MS assays. They also ran the samples on a GC-MS column shared by others in the department, with the original stock column, instead of using a dedicated instrument, like that which is used at Enthalpy Analytical.
Marquette asked the Journal Sentinel to have Enthalpy to verify their results, not the other way around. Enthalpy has years of experience with their methods, which are validated, and have run these tests 1000s of times. Not only this, but Enthalpy verified their own results with two methods: GC-MS and derivative HPLC, and those two methods got matching results between them for each sample given that per ml and per gram can differ by up to 30% depending on sample density. Enthalpy Analytcial’s methods are validated, accredited, and they run these assays over 500 times a month. Dr. Reid has run a
total of 5 samples
ever, all for this article, and with a method not validated for e-liquid samples, but instead is recommended for the analysis of diketones in beer. I have not spoken with him myself, but I am told by Gene Gillman (Director of Enthalpy Analytical) that Dr. Reid is a very nice and reasonable man, and he is
not happy with how this story ended up, and certainly does not agree that his method is "superior", no matter what the unspecified "experts" said in the article. Enthalpy is currently working with Dr. Reid to understand the differences between the methods. This is how science is done!
Here are the results of the two labs:
The units for the data are given as “ug/mL”. I will assume she means “μg/mL”, micrograms per milliliter, since ug is not an actual unit. Did Ms. Rutledge or the Sentinel’s editor not know this? Given the focus on small LOD differences between the two labs, I am troubled by this unit error.
Enthalpy was asked to run
for verification 3 samples. They ran and got, using two different methods, reproduced results for Foghorn Randy, Matik Pomegranate, and Radioactive Vapes Grape. The last two were measured as ND (non-detects, zero within the method's LOD), while the Randy had 61.8 μg/mL DA and 207 μg/mL AP, again, results matching between two methods. This is the GOLD STANDARD for such a test: to have results matched between two different validated methods, in this case GC-MS and derivative-HPLC. These results are somewhat different than Marquette's first time ever doing a DA or AP test, and with an unvalidated method, and without an internal standard, let alone a 5-point calibration curve.
But according to Raquel, who is clearly not an analytical chemist, none of that matters at all. Marquette claims a slightly lower LOD, so it is superior, end of story. Forget verification. Forget years of intense and voluminous experience working out dilutions and flavor-matrix problems. Forget published methods for testing DA and AP in e-liquids in peer-reviewed journals. Forget that LOD and LOQ are practically
invalid terms if the method is not validated, and you have essentially zero experience running it.
It should also be noted that while Ms. Rutledge seems to be trying to insult Enthalpy as being inferior to Marquette, which is not based in any data whatsoever, and makes zero sense, it was Enthalpy that helped first bring this problem of diketones into the open, and was the first to publish peer-reviewed validated tests to determine their concentrations in e-liquids. They also were the ones that measured DA in cigarette smoke (acknowledged in Pierce, J.S., et al., Crit. Rev. Toxicol., 2014), which is on average about 100x the exposure from vapor from positive-testing e-liquids, something that Raquel evidently does not find important. It should also be noted that there are no data to suggest that levels of DA or AP in e-liquids below 1-5 μg/mL have
any negative health impact.
Now, based on the implications of the article, Matik and Radioactive Vapes think they have been selling two DA- and AP-positive e-liquids, when these compounds are almost certainly
not present at all.
Now Dr. Reid is in, for an ethical chemist, the very uncomfortable position of having essentially preliminary data of an assay still in development being labeled as superior to an established commercial lab using two validated methods for verification of results, something Dr. Reid himself does not agree with.
Now an established commercial analytical lab that has specialized for years in e-liquids and tobacco products, and has been a long trusted vanguard for accuracy in testing for AEMSA for this very problem, and who was integral to first publishing this DA/AP problem, is being attacked because their validated and verified results do not match that of a lab that has no experience with this assay, using an unvalidated and uncalibrated method. How can this possibly be? What is the purpose of this complete lack of regard for rigorous analytical chemistry?
Now this story is being carried by other outlets with a new title that is
meant to slander Enthalpy Analytical, Inc. If these results were part of a manuscript submitted to a scientific journal for publication, and the conclusions Ms. Rutledge has put forth were the conclusions of the paper, it would be rejected upon peer-review based on profoundly faulty logic and a complete lack of understanding of fundamental analytical chemistry, starting with the use of incorrect units.
If Ms. Rutledge wanted the industry to solve this problem, she has gone about it completely wrong. With this article a major validated and verified tool in the very solving of this important problem, which Ms. Rutledge should
support, is being attacked with accusations which do not stand at all to rigorous and educated scientific scrutiny, and with conclusions that
all scientists involved with this study
do not agree with.
In my professional opinion, as PhD chemist who has published in the electronic cigarette field in peer-reviewed journals, and has valued, trusted and rigorous collaborations with Enthalpy on these very matters, this story needs to be either drastically rewritten to reflect the
actual science, or else removed entirely, including from other outlets that are now spreading this ignorant and pointless attack. Either way, a public apology should be given by Ms. Rutledge, if not the Milwaukee Journal Sentinel.
Ms. Rutledge, if you really want to help this important cause and promote truly accurate testing, which is critical if we ever hope to solve this, do the right thing here. Rewrite this story, or have it taken down.