Statistical variation of nerve conduction velocity. An analysis in normal subjects and uraemic patients.

The maximal nerve conduction velocity (NCV) of 6 sensory nerves (N. radialis, suralis and peronaeus superficialis, bilaterally) was determined in 88 normals and 59 uraemic patients. When the effect upon nerve conduction velocity of age and temperature was removed through partial correlations, it became possible to verify the negative correlation between the height of the patient and NCV. This explained as much as 35% of the remaining variance in males. There were also other significant differences in the nerve functions between the sexes. A set of "statistical corrections" (grouping of NCV's according to the subject's sex, computing of the average NCV's of an individual, and height-, age- and temperature normalization of single or mean NCV values) substantially decreased the variability of NCV. At the same time, the sensitivity of NCV measurement in the early diagnosis of nerve dysfunction caused by uraemia was increased. It is suggested that the results of NCV measurements should not only be reported as rough, absolute data, but also as normative values, taking into account the known physiological determinants.