Is this study on MRSA inhibition by essential oils legitimate?
August 1, 2011 4:47 PM Subscribe
Is this study about the inhibitory effect of essential oils legitimate?
I'd link to more than the abstract, but I only have access to it by remotely accessing my university library's databases. I found it through wikipedia's entry on MRSA, but I know that it was funded by an essential oil company. Are the methods used typical for this kind of study? Are the results significant?
I'd link to more than the abstract, but I only have access to it by remotely accessing my university library's databases. I found it through wikipedia's entry on MRSA, but I know that it was funded by an essential oil company. Are the methods used typical for this kind of study? Are the results significant?
Best answer: I know that it was funded by an essential oil company.
It was performed by the essential oil company. The first two authors, including the corresponding authors are affiliated with the company, and the second author is the founder and president of the company. This is a conflict of interest of an unusual magnitude. The idea of treating essential oils as a uniform class is somewhat dubious chemically, and I usually think of it as a marketing term that makes me cringe.
As far as the particular results showing in vitro inhibition of MRSA by various essential oils, I'm a little concerned by the lack of even a single illustrative image of an actual disk diffusion example. But I looked up a couple of similar papers and it seems like it is common to report results for this kind of assay only in table form.
The study also lacks an adequate positive control. They only compared against methicillin, which, of course, MRSA is resistant against. It would have been really easy to compare against ethanol, but they didn't. So this doesn't really give you a picture of how well essential oils perform against things that we already know to work.
Even assuming that the study was performed correctly, their conclusions are a big stretch, although it is couched here, at least, in appropriate weasel words1 such as "may include" and "potential use":
The gold standard for showing that a particular substance has a therapeutic effect is a randomized controlled trial. That doesn't say that the methods used here were incorrect, although the problems I identified with the study above probably go some way to explaining why it was published in such a minor journal. But it does mean that the study doesn't mean what many people are going to use it to mean—it provides little evidence on the therapeutic use of essential oils.
1 Lest there be any doubt, overreaching conclusions couched by weasel words is not a problem unique to this paper.
posted by grouse at 5:36 PM on August 1, 2011 [7 favorites]
It was performed by the essential oil company. The first two authors, including the corresponding authors are affiliated with the company, and the second author is the founder and president of the company. This is a conflict of interest of an unusual magnitude. The idea of treating essential oils as a uniform class is somewhat dubious chemically, and I usually think of it as a marketing term that makes me cringe.
As far as the particular results showing in vitro inhibition of MRSA by various essential oils, I'm a little concerned by the lack of even a single illustrative image of an actual disk diffusion example. But I looked up a couple of similar papers and it seems like it is common to report results for this kind of assay only in table form.
The study also lacks an adequate positive control. They only compared against methicillin, which, of course, MRSA is resistant against. It would have been really easy to compare against ethanol, but they didn't. So this doesn't really give you a picture of how well essential oils perform against things that we already know to work.
Even assuming that the study was performed correctly, their conclusions are a big stretch, although it is couched here, at least, in appropriate weasel words1 such as "may include" and "potential use":
Application of these results may include the potential use of essential oils as an alternative therapy for various diseases sustained by S. aureus MRSA.In the paper they suggest that essential oils might be "administered either orally or dermally." So, hey, it might work. But there's very little data to suggest that it would. Let's compare ethanol. Ethanol inhibits MRSA quite well on contact, but if oral administration of ethanol were enough to kill it off, then we'd just send MRSA patients to the local whisky bar. Unfortunately, the amount of ethanol that is sufficient to kill one's MRSA infection is quite possibly enough to kill the patient first. The same may be true of essential oils.
The gold standard for showing that a particular substance has a therapeutic effect is a randomized controlled trial. That doesn't say that the methods used here were incorrect, although the problems I identified with the study above probably go some way to explaining why it was published in such a minor journal. But it does mean that the study doesn't mean what many people are going to use it to mean—it provides little evidence on the therapeutic use of essential oils.
1 Lest there be any doubt, overreaching conclusions couched by weasel words is not a problem unique to this paper.
posted by grouse at 5:36 PM on August 1, 2011 [7 favorites]
Best answer: It's known that a number of common essential oils have weak antimicrobial activity, but I'm not familiar with their safety profiles. At high enough concentration, I do know that most phenolic compound will harm cells, which is why they're commonly used as disinfectants (ie. hexachlorophene, chloroxylenol).
Table 3a in that paper breaks some of the oils down into their component mixtures, which is what you really need. This paper (DOI:10.3390/molecules1511753, open access) contains the activities of pure essential oil components, whereas this one can be used to convert disc radii into the more commonly used ug/mL metric (doi:10.1111/j.1365-2672.2007.03353.x), which will allow you to do straight comparisons with known antibiotics (disc diffusion is good for a quick active/not active determination, but doesn't seem to be usually used to quantify activity).
As for significance, I wouldn't expect any essential oil to become a drug in the near future. At best they might be stocked in as an "alternative" disinfectant, but ingesting a significant quantity is just as likely to kill you as the bacteria.
PS. For fun, similar results with phenolic compounds from marijuana (link).
posted by Orange Pamplemousse at 11:12 AM on August 2, 2011
Table 3a in that paper breaks some of the oils down into their component mixtures, which is what you really need. This paper (DOI:10.3390/molecules1511753, open access) contains the activities of pure essential oil components, whereas this one can be used to convert disc radii into the more commonly used ug/mL metric (doi:10.1111/j.1365-2672.2007.03353.x), which will allow you to do straight comparisons with known antibiotics (disc diffusion is good for a quick active/not active determination, but doesn't seem to be usually used to quantify activity).
As for significance, I wouldn't expect any essential oil to become a drug in the near future. At best they might be stocked in as an "alternative" disinfectant, but ingesting a significant quantity is just as likely to kill you as the bacteria.
PS. For fun, similar results with phenolic compounds from marijuana (link).
posted by Orange Pamplemousse at 11:12 AM on August 2, 2011
Response by poster: Thanks so much. I may try to edit this off of the wikipedia page.
posted by sunnichka at 5:24 PM on August 2, 2011
posted by sunnichka at 5:24 PM on August 2, 2011
« Older Gmail interface with Exchange data | Do you think shelves instead of kitchen cabinets... Newer »
This thread is closed to new comments.
posted by gramcracker at 4:51 PM on August 1, 2011 [1 favorite]