Estimating Reference Limits for Motor Conduction Block Using Latency Adjustment and a Modified MeRef Model

Current thresholds for motor conduction block on nerve conduction studies are primarily based on expert opinion and fail to consider individual patient- or nerve-specific characteristics. In this article, we aimed to improve diagnostic accuracy for identifying partial motor conduction block and determine the influence of latency difference (LD) between proximal and distal stimulation on amplitude decay in unaffected nerves.

We developed the multivariate extrapolated truncated fit model to establish reference limits for amplitude decay depending on LD using existing nerve conduction studies data and compared these with existing literature values and with published thresholds for conduction block.

A total of 67,266 investigations of the forearm or lower leg segments of the median, ulnar, peroneal, and tibial nerves were included. The mean multivariate extrapolated truncated fit upper limits (model +2SD) for amplitude decay in nerves with normal distal amplitude/decreased distal amplitude were: Median 14.3%/24.1%, Ulnar 17.5%/34.4%, Peroneal 27.7%/41%, and Tibial 46%/57.1%. For the four nerves, we observed 1.7% to 5.1% increases in amplitude decay per ms increase in LD. Contrary to the multivariate extrapolated truncated fit reference limits, existing thresholds for conduction block produced inconsistent positive rates between nerves with a high risk of false positive and negative results.

Amplitude decay on nerve conduction studies was dependent on LD, and the multivariate extrapolated truncated fit model showed promise as a tool to create reference limits from large data sets. Reference limits adjusted for LD could improve diagnosis of inflammatory polyneuropathies and other conditions with conduction block.