Sensory nerve conduction in the caudal nerves of rats with diabetes 1 Condução nervosa sensorial no nervo caudal de ratos com diabetes experimental

Purpose: To investigate sensory nerve conduction of the caudal nerve in normal and diabetic rats. Methods: Diabetes was induced in twenty 8-weeks old Wistar male rats. Twenty normal rats served as controls. Caudal nerve conduction studies were made before diabetes induction and the end of each week for six consecutive weeks. The caudal nerve was stimulated distally and nerve potentials were recorded proximally on the animal’s tail using common “alligator” clips as surface electrodes. Results: After induction, nerve conduction velocities (NCV) increased slower in the diabetic than in the control group. Sensory nerve action potentials (SNAP) conduction velocities increased slower in the diabetic than in the control group (slope of regression line: 0.5 vs 1.3m/s per week; NCV in the 15th week = 39±3m/s vs 44±4m/s). Tukey’s tests showed differences between groups at the 11th, 13th and 15th weeks old. From the 10th week on, SNAP amplitudes increased faster in the diabetic than in the control group (slopes of the regression line: 10 vs 8μV per week; SNAP amplitudes in the 15th week: 107±23μV vs 85±13μV). Differences at the 12th, 13th and 15th weeks were significant. Conclusion: In diabetic rats nerve conduction velocities were slower whereas amplitudes were larger than in normal rats.

The objective of this work was to investigate sensory nerve conduction of the caudal nerve in normal and diabetic rats.

Methods
Forty Wistar rats were used (Rattus norvegicus), male albino rats with 8-weeks old, from the Nutrition Department Animal Lab Center of the Federal University of Pernambuco.Animals were kept in collective plastic cages (maximum of 5 animals/cage), under a temperature of 23±1ºC, an inverted 12h day/night cycle, with free access to water and food (Labina).Animals were randomly divided in two groups of 20 animals: The Control Group, diabetes free and the Diabetes Group, with nontreated experimentally induced diabetes.Diabetes was induced by a single intraperitoneal injection of 60mg/kg streptozotocin (Sigma Chemical Co., USA) diluted in buffer solution (sodium citrate, 10mM, pH 4.5) after 12 hours of fasting 23 .Animals in the control group received the same buffer solution without streptozotocin.After 30 minutes both groups were fed normally.Blood glucose was measured 3 and 7 days after induction, with blood samples collected from the tip of the tail after 12 hours of fasting.Only animals with blood glucose above 200mg/dL (Glucosymmetry AccuCheck -Roche) were admitted in the Diabetes Group.Both groups received a weekly blood glucose check up.
Methods applied to handle and care for animals are in accordance with the National Institute of Health Guide for Care and Use of Laboratory Animal standards and were approved by the Ethics Committee for Animal Experimentation of the Federal University of Pernambuco.

Electrophysiology
Caudal nerve conduction studies were made before diabetes induction and the end of each week for six consecutive weeks.After anesthesia with a intramuscular injection of a mixture of Xylazin Chlorhydrate (Rompum) and Ketamin (Ketalar), 0.2ml/100g 19 , animals were positioned ventrally for their tail to be completely loose and the tail was clean and fat free with alcohol at 70%.
An orthodromic technique was used 19 : The caudal nerve was stimulated distally (duration: 0.1ms; intensity: 10-20µA, supramaximal) and the sensory nerve action potentials (SNAP's) were recorded proximally (high-pass filter: 10Hz; low-pass filter: 10kHz) on the animal's tail.No movement of the tail was noted during stimulation.A Racia-Alvar Centor electromyograph was used.The SNAP parameters studied were peak-to-peak amplitude (µV) and nerve conduction velocity (m/s), derived from the onset latency (ms).Measurements were taken 4 times and the accepted result was their arithmetic average.Stimulation, recording and ground electrodes were identical standard "alligator" clips 19 .The distance between the stimulating cathode and the non-inverting recording electrode was 8cm, whereas the distance between the pair of stimulating and recording electrodes was 3cm.The ground was connected to the pair of alligator clips located between the stimulation and recording electrodes.The springs of the alligator clips were removed to prevent excessive compression of the tail.
Due to the almost linear relationship between the temperature of the tail and the nerve conduction velocity 20 , before each recording the temperature of the tail was maintained at 31-32ºC and confirmed before each recording with an infrared thermometer (Doc Thermo; Comdek Industrial Corp.).A dichroic lamp was used for heating.

Statistical analysis
Data were summarized by mean and standard deviation.The statistical significance of the differences was tested with mixedmodel analysis of variance (ANOVA).Post-hoc comparisons were made by Tukey test.The critical p for both tests was 0.05.

Results
After two weeks of the induction period the animals in the experimental group showed increased levels of blood glucose, whereas those in the control group had normal glucose levels (Table 1).SNAP conduction velocities of diabetic and control animals are shown in Figure 1.There was no significant difference between them at eight weeks of life (pre-induction): 36±2m/s versus 34±3m/s, respectively.After induction, SNAP conduction velocities increased slower in the diabetic than in the control group (slope of regression line: 0.5 versus 1.3 m/s per week; NCV in the 15 th week = 39±3m/s versus 44±4m/s).Post hoc tests (Tukey) showed significant differences between groups at the 11 th , 13 th and 15 th weeks.Figure 2 shows SNAP amplitudes of diabetic and control animals.At the 8 th week (pre-induction; baseline) the SNAP amplitudes were not different between the two groups (33±7µV versus 29±6µV).From the 10 th week, SNAP amplitudes increased faster in the diabetic than in the control group (slopes of the regression line: 10 versus 8µV per week; SNAP amplitudes in the 15 th week: 107±23µV versus 85±13µV).Differences at the 12 th , 13 th and 15 th weeks were statistically significant (Tukey).Rats with diabetes lost weight during the experiment, whereas those in the control group gained weight (Figure 3).In the 15 th week, the average weight of the diabetic rats was about 40% lower than the weights of the control rats.

Discussion
The rat caudal nerve has good characteristics for studying nerve conduction using non-invasive techniques.Its superficial trajectory allows easy stimulation and recording of potentials with surface electrodes 21 and its long trajectory (10-15cm) allows accurate measurement of distances 18 .
In our data SNAP conduction velocities increased slower with age in the diabetic rats than in the normal rats.This was an expected result.Several authors obtained similar results studying the caudal nerve 2,4,5,20 or other nerves 3,6,7 .The slower conduction velocity is probably due to an abnormal development of the myelin sheath in the peripheral nerves of the diabetic rats.
Caudal nerve SNAP amplitudes increased faster in the diabetic rats than in the normal rats.This was an unexpected result.We expected a decrease in the rate of growth of the SNAP amplitude in the diabetic rats, due to the axonal loss and temporal dispersion of SNAP related to the diabetic polyneuropathy.Zochodne and Nguyen 22 reported larger caudal nerve SNAP amplitudes in diabetic rats compared to non-diabetic rats.However, they did not discuss this result nor state if this difference was statistically significant.
The higher amplitude of the SNAP in the diabetic rats could be caused by a lower temperature of the animal's tail.Low temperature is known to cause SNAP amplitude to increase and nerve conduction velocity to fall 20 .However, we monitored the temperature of the tail and kept it fairly constant.
Most probably the reason for this finding was the smaller development of the tail diameter in the diabetic rats, related to those in the control group.In a volume conductor, the amplitude of the SNAP (a quadripole source) decreases in the inverse ratio of the cube of the distance between the nerve and the registry point 23 .Therefore, the smaller diameter of the tail, and consequently the shorter distance between the caudal nerve and the recording electrode in the diabetic rats, would result in larger SNAP amplitudes.
As we did not measure the diameter of the animal tails, these variations can be indirectly estimated by variations in the animal's weight.For any linear dimension of a body of any shape is proportional to the cubic root of its volume 24 .Because the weight of a body is proportional to its volume, the diameter of the rat's tail is proportional to the cubic root of the rat's weight.Our rats with diabetes lost weight during the experiment, whereas those in the control group gained weight.Thus, it is reasonable to suppose that the diameter of the tails, and consequently the distances between the caudal nerve and the recording electrodes, were significantly smaller in the diabetic rats than in the control rats.Authors that use needle type electrodes for recording [2][3][4][5][6]12,14,15 did not observe this phenomenon and the amplitudes behaved as expected, probably because the recording electrode was always placed close to the nerve being studied, independently of the weight of the animal.

Conclusion
The nerve conduction in diabetic rats was slower whereas amplitudes of action potentials were larger then compared with normal rats.

FIGURE 3 -
FIGURE 3 -Mean (standard deviation) of the weight (g) by age (weeks) of diabetic (filled circles) and control (open circles) rats.

TABLE 1 -
Mean (standard deviation) of blood glucose levels (mg/100ml) by age (weeks) of diabetic and control rats