A study carried out at the Royal Veterinary College, London, found that, when assessing the severity of the lameness, observers were influenced by whether they knew that a nerve block had been performed.
The diagnosis of all but the most obvious causes of lameness requires the use of additional tests such as nerve blocks. The veterinarian assesses the severity of the lameness before and after anaesthetising specific areas of the horse’s limb.
Interpreting nerve blocks is not always straightforward. The anaesthetic may spread to affect a wider area than was intended. On the other hand, a block may fail to desensitise the target area because of anatomical variations in the location of nerve branches.
Another possible source of error is the assessment of the response to the nerve block. The clinician may already have their suspicions about the site of the problem. Does that influence their assessment of the lameness before and after the nerve block? A recent study has been investigating.
Ms Marie Arkell and colleagues at the Royal Veterinary College, London, used a series of video recordings of lame horses taken both before and after a nerve block. The videos were shown to different observers - including orthopaedic experts, general clinicians and final year students.
Initially, the observers examined all the videos at random, without knowing which horse had been nerve blocked. They gave each horse a lameness score, graded from 0 (sound) to 10 (non-weight bearing). Then they repeated the process with each horse’s recordings paired together, so that the recording taken after the nerve block was viewed immediately after the one recorded before the horse was blocked.
Overall there was a greater difference in the two lameness scores when the observers knew that a nerve block had been performed. On average, once the observer knew that the horse had been nerve-blocked, the difference in lameness score increased by 0.4.
Individual observers were quite consistent in assessing the grade of lameness. They showed an average difference of 0.6 grades when examining the same horse on two occasions.
There was good agreement between the assessments of the orthopaedic experts, as might be expected. The lameness grades given by the non-experts and final year students showed more variation.
According to the researchers, this bias and variation is most likely to be significant in horses that are only slightly lame. Obviously in cases of moderate or severe lameness when there is a marked change in degree of lameness any effect of bias would be negligible.
They suggest that a video recording be taken to supplement the clinical notes, especially when investigating low-grade lameness. This would allow a assessment by an unbiased observer at a later stage.
For more details see:
Evidence of bias affecting the interpretation of the results of local anaesthetic nerve blocks when assessing lameness in horses.
M Arkell, RM Archer, FJ Guitian, SA May.
Veterinary Record (2006) 159, 346 - 349