A Comparison of Electrophysiological Tests for the Early Diagnosis of Diabetic Neuropathy.

A Comparison of Electrophysiological Tests for the Early Diagnosis of Diabetic Neuropathy Pastore C, Izura V, Geijo-Barrientos E, Dominguez JR (Seccion de Endocrin ologia y Nutricion and S. de Neurofisiologia Clinica (EMG), Hospital Universitario de San Juan, San Juan, Alicante, Spain; Departamento de Fidiologia and Instituto de Neurocienciascsic, Universidad "Miguel Hernandez" de Elche, Alicante, Spain), Muscle Nerve. 1999;22:1667-1673.

The purpose of this study was to describe and compare nerve conduction tests of patients with diabetes mellitus type I (DM 1) and diabetes mellitus type II (DM 2) over a period of time. In addition, the study evaluated these tests as early indicators of subclinical diabetic polyneuropathy.

Subjects included 99 patients with diabetes mellitus seen in the Endocrinology Unit of the San Juan University Hospital. The average age of the subjects was 39.8 years, and the average duration of diagnosed illness was 10.6 years. Using the criteria of the American Diabetes Association, 60 subjects (36 women, 24 men) were classified as having DM 1, and 39 (26 women, 13 men) were classified as having DM 2.

All subjects were examined at least twice within a 4-month period. In addition to biochemical and hematological tests, patients had a neurological examination detailing reflexes and sensitivity to pain, touch, and vibration. The 3 criteria used to define diabetic polyneuropathy were depressed tendon reflexes in the limbs or absent ankle jerks, reduced vibratory sense below the knees, and reduced light touch sensory discrimination in the legs. If a subject had 2 of these 3 conditions and did not have brisk reflexes, he or she was classified as having diabetic polyneuropathy.

The subjects also hadconventional sensory and motor nerve conduction studies of the median, ulnar, radial, tibial, and sural nerves. Median and ulnar sensory studies were performed on one arm, and radial and sural sensory studies were performed on both extremities. The sensory nerve action potential (SNAP) amplitudes of the radial and sural nerves were quantified using 2 different measures: one from the baseline to the peak of the negative wave, the other from the negative peak to the following positive peak. The sural/radial ratio for each measure was calculated. After completion of motor nerve conduction studies, the left tibial nerve at the malleolus was stimulated, and the F-wave latency was recorded at the left abductor hallucis muscle.

Data from 172 control subjects were also recorded to define the limits of normality and to correct for age and height. These volunteers were recruited from hospital laboratory staff, students, or patients referred to the Endocrinology Unit with no evidence of peripheral neurological disease.

The unpaired Student t test was used to compare different nerve conduction parameters, and the Pearson product moment correlation coefficient was used to indicate the correlation between variables. F-wave latency values were corrected for age and height; sural SNAP amplitude and sensory nerve conduction velocities were corrected for age only.

Results showed that 29.3% of the subjects with DM 1 or DM 2 had abnormal sural nerve SNAP amplitudes and that 29.8% had abnormal sensory conduction velocity values. These numbers were similar to the percentage of subjects diagnosed as having diabetic polyneuropathy by clinical criteria (31.3%). Abnormal values of minimal tibial F-wave latency were found in 36.4% of the subjects, whereas 51.0% had low sural/radial amplitude ratios. No significant correlation was found between the duration of the illness and the electrophysiological parameters in subjects with DM 2. Subjects with DM 1, however, did have a strong correlation between sural/radial amplitude ratio and duration of disease when the duration was more than 6 years. Sural SNAP amplitude, sensory conduction velocity, and the inverse of the minimal tibial F-wave latency also decreased with disease duration, but they were significant only in subjects with DM 1 whose disease duration was greater than 9 years.

The authors concluded that, in patients with either DM 1 or DM 2, it is reasonable to consider the simple ratio between sural and radial sensory amplitude as a more sensitive early diagnostic test for diabetic polyneuropathy than sural SNAP amplitude alone or minimal F-wave latency. They also stated that the simple ratio was also found to be the earliest electrophysiological parameter to detect nerve pathology in subjects with DM 1.

Nancy Talbott, PT University of Cincinnati Cincinnati, Ohio