STEVEN H. HOROWITZ, M.D.
To the editor: Graf and associates (1) conclude from measurements of nerve conduction velocities that diabetic control improves nerve function and thereby support the hypothesis that diabetic neuropathy has a metabolic component. Although these conclusions may be correct, they are based on inadequate and misleading data and represent a gross misuse of statistics.
Of five motor and sensory velocities studied, one (median motor) showed statistical improvement, 3.7 m/s (8% of the pretreatment value), after 12 months of therapy. Practically, repeated differences of less than 5.0 m/s on the same nerve are considered to be within the test's range of variability.
HOROWITZ SH. Diabetes Control and Neuropathy. Ann Intern Med. 1981;95:654–655. doi: 10.7326/0003-4819-95-5-654
Download citation file:
Published: Ann Intern Med. 1981;95(5):654-655.
Copyright © 2018 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use