Continuity with filaments

Von Frey filaments may not be perfect, but they are tried and tested and researchers know from previous studies what mechanical thresholds to expect. The force measurements may not be correct (in absolute terms) but they are reproducible and, as long as the method detects a change in the treatment group but not in the control group, there may be little incentive to upgrade the measurement method for future studies.

Topcat is sympathetic to this view and our in-house von Frey validation system, MouseCal, addresses the issue of reproducibility with both filament and EvF measurements. We (Dr Polly Taylor and Dr Michael Dixon) began measuring thermal nociceptive thresholds in cats in 1998, using our own thermal stimulus. By 2008, when Topcat Metrology was formed, we had developed and sold three generations of thermal probes in our quest for absolute accuracy of the threshold temperature without tissue damage. Similarly, for mechanical threshold testing, we had investigated the effect of probe diameter and the influence of loading rate.

In both these fields, we made it a priority to validate new designs against previous generations, to understand the differences in the data and to make this information freely available to the scientific world. This allowed new studies to be compared realistically with older work and has increased the value of our global dataset enormously over the last ten years. Increasingly, thermal and mechanical nociceptive testing in the target species using Topcat’s equipment is seen as a milestone in the registration process of new analgesic drugs; we are the only supplier in the world of wireless thermal and mechanical nociceptive testing systems for animal use.

We are firmly committed to the same approach for MouseMet and RatMet. Our own validation studies have shown us that there are intrinsic differences betweeen the von Frey filament method and an analog force transducer; even if both methods were executed perfectly, one should not necessarily expect to measure the same mechanical threshold force. This doesn’t mean that the two cannot be compared; it just means that allowance must be made for the differences.

Here, in summary, is our current understanding of the reasons for differences between von Frey filament data and that from an analog force transducer (commonly known as Electronic von Frey, or EvF). We believe the comments to be applicable to all systems, not just our own, although those with inappropriately high force ranges are likely to incur additional errors. We do not claim to have the full picture yet. Please, if you have a comment or some relevant data that you would like to share with us, get in touch.

  • Each von Frey filament has a unique tip diameter, whereas the probe on an EvF is the same irrespective of the force that it applies. It is well accepted that threshold force increases linearly with tip diameter (here’s a recent Topcat poster) and allowance should therefore be made for this when attempting to make a comparison of one data point (perhaps the baseline).
  • The area and shape of the von Frey filament contact patch change drastically as it buckles. This makes the comparison even more difficult.
  • The variation in tip diameter  “stretches” the von Frey filament force scale, lowering treatment thresholds with hyperalgesia unrealistically. For quantification of this effect, click here. (It needs thinking about, so this page is a pdf that you can print out and scribble on)
  • The up-down method, commonly used to analyse von Frey filament data, is prone to errors if used incorrectly. Here’s a recent Topcat poster in which we quantify the possible errors.
  • Some electronic von frey systems (not Topcat’s) consist of a conventional strain gauge force transducer mounted in a hand held “wand”, with a probe on the measurement side to contact the rodent’s foot. This probe (and its attachment) may have a significant mass compared to the forces being measured. This will mean that the transducer will register a force if the probe is subjected to significant acceleration along its axis, such as can be generated by waving the probe around in air without ever going near the rodent. The force indicated is simply a result of the equation Force=Mass*Acceleration. To accelerate the probe mass, you need a force. (Click here for a paper written by Dr MJ Dixon while studying the effect at the National Physical Laboratory in the 1980s).

In Summary:

MouseMet (or RatMet) may not give you the same baseline threshold as your von Frey filaments and it may indicate a reduced treatment effect compared to your previous studies. But the measurements of mechanical threshold force will be more accurate and just as reproducible (we know what the spread of your data should be and we’ll coach you to achieve a good standard deviation with the early baselines). They will also be quicker (and therefore cheaper) to generate, and less dependent on the operator.