Have electronic von Frey instruments been found effective for testing mechanical sensitivity in mice? Which models are the most effective?
tl;dr: People use them, yes. They’re accepted in the literature. But each eVF is different and has pros and cons. And they measure different things than traditional VFs. They may lack sufficient sensitivity to see changes on the low end of the PWT.
Thanks for the question. Some labs and papers do use electronic von Frey apparatus. Traditional von Frey monofilaments are by far more common (I’d say 80% of studies), but eVFs are used and accepted. Not all eVFs are the same and depending on who you talk to, some are better than others. I think there are a lot of advantages to using an eVF including:
- linear continuous measurement
- more objective outcome (loss of contact from the probe)
- more information (rate of force, latency to response, etc.)
Concerns are that they are not as sensitive as VFMs (monofilaments). Many of the commercially available eVFs are made with force ranges for Rats rather than Mice.
The available eVFs are:
I may have missed some, but these are the most frequent ones.
My lab has the IITC and we had the MouseMet for a while. The IITC one works, but the probe is stiff and the force range, IMO, is too large for mouse. It works well for rats I’ve heard.
When I first started grad school, I was really into improving methods, and so I wanted to get an eVF. I found the MouseMet (MM) one and was very impressed by the thoughtful design. If you look at their site, you’ll see that the inventors understand what pain researchers are trying to measure and what the common pitfalls and disadvantages of VFMs are.
I used MM for about 1.5 years. There was a lot to like about it, but ultimately, I felt that it was not sufficiently sensitive to see smaller differences in PWTs at the lower end of the range (say, a sensitive SNI mouse or CFA). I think if it could be more sensitive in that lower range, it’d be a winner. Also, something to keep in mind for all these eVFs is that they differ from VFMs in that they apply a dynamic force rather than a static punctate force. So a VFM reaches a max force when it bends and does not increase further, whereas the MM and the UG Dynamic apply a changing force until the mouse withdraws. That likely affects the PWT. Indeed, if you look at the numerical force values for eVFs vs. VFMs, they’re widely divergent ( Avg 1g for VFMs vs 2-4 g for eVFs, depending on model). They talk about all these things on their site (which is one reason why I like their approach)
For me, in the end, I switched back to VFMs. They get the job done and they’re widely accepted. There are a lot of things to criticize about the assay in general (what actually does a paw withdrawal even mean?). But until the VF assay is abandoned, one has to pick the least bad tool, and VFMs are probably it, for now… A much better solution would be to come up with better, more informative assays for pain, but that’s a much larger challenge.
@thicunha uses an eVF in his lab, I believe. Hopefully he and anyone else who uses eVFs will chime in.
This is very helpful, thank you for your input. I am measuring PWTs in mice with elevated sensitivity, so I am looking for an eVF that would be sensitive at the very low end of the range. It looks like some of these instruments come with special tips they claim are sensitive enough for mice, such as the “Supertips,” on the IITC model. Have you ever tried any of these?
I haven’t used, but I’m not sure I believe it. @thicunha - Have you used the IITC eVF with the Supertips?
@snlanger If you’re in the market for an eVF, consider the MouseMet. I looked at the specs, and since the time I used it, they have reduced the force range to 0.1g from 0.5g. That’s really important. If I had more time, I’d try it out again. The price is right too. Looks about ~$2500. The thermal device they made looks sweet too.
Btw: I don’t get paid to promote them I just like the company and the people and think their stuff is the most advanced of eVFs and they’ve thought about the relevant variables.