J. Endocrinol. Invest. 15.279-282,1992

Changes in central and peripheral nervous system function during hypoglycemia in man: an electro-physiological quantification G. Tamburrano*, N. Locuratolo*, G. Pozzessere**, O. Lostia***, S. Caiola***, E. Valle**, F. Bianco**, A. Giaccari*, and P.A. Rizzo** * Cattedra di Endocrinologia I, ** Istituto di Clinica delle Malattie Nervose e Mentali, Universita di Roma "La Sapienza". and *** Istituto Superiore di Sanita, Roma, Italy

ABSTRACT. We measured somatosensory evoked potentials (SEP) in normal subjects during acute (group A) and moderately prolonged (group B) hypoglycemia. We considered the following parameters: peripheral conduction velocity (wrist-Erb CV), conduction time (CT) between brachial plexus and the cervical cord (Erb-Nd and central CT from the cervical cord/lower brainstem lemniscal pathway to the cortex (N 13-N 20 ). In group A, the electrophysiological parameters did not change significantly

throughout the study. In group B, mean N13-N 20 CT increased from a basal values of 5.82 ± 0.11 to 6.22 ± 0.11 msec at 105 min (p < 0.02) and 6.33 ± 0.11 msec at 120 min (p < 0.05). This study indicates that neither acute nor moderately prolonged hypoglycemia influence the peripheral nerve function in normal subjects and provides evidence that hypoglycemia as low as 2.4 mmol/L, lasting more than 60 min, can significantly increase the conduction time of central somatosensory pathways.

INTRODUCTION

plored using this technique, from the peripheral skin to the cerebral cortex, makes it vulnerable at several levels in a variety of pathological conditions (9). The aim of our study was to investigate the SEP behavior in man, following median nerve stimulation at the wrist, during acute and moderately prolonged insulin-induced hypoglycemia. Median nerve stimulation was chosen because of the more frequent involvement of the upper limb in the syndrome of peripheral polyneuropathy associated with islet cell tumors and hypoglycemia (7).

Under normal conditions, the brain and the peripheral nervous system depend on glucose as the major substratum for energy production (1, 2). During the past few years, several studies examined the response to hypoglycemia and showed neurophysiological as well as neuropsychological changes (3,4). Insulin induced hypoglycemia has been shown to produce acute axonal degeneration in both non diabetic and diabetic (5) animals, and to alter the anterograde fast components of axonal transport in rats (6). Furthermore, a syndrome of peripheral polyneuropathy associated with hypoglycemia secondary to insulinoma has been reported in several patients (7). Somatosensory evoked potentials (SEP) represent the electrophysiological response of the nervous system to the sensory stimuli of nerves in both lower and upper limbs. SEPs are mediated by rapidly conducting fibers, predominantly I-a large diameter myelinated fibers (8). The remarkable length of the nervous pathways ex-

MATERIALS AND METHODS Twelve normal healthy subjects, all within 10% of their ideal body weight (according to the Metropolitan Life Insurance Co. tables, 1959), were studied. They did not have a family history of diabetes or neurological disorders and were divided into the following groups: Group A: six subjects (3 men and 3 women, aged 26-30 yr) in whom an acute hypoglycemia was induced by an injection of regular insulin as iv bolus (0 1 U/kg). Samples for the determination of glucose concentration (BG) in arterialized blood were withdrawn every five min from a catheter inserted into a heated hand vein, immediately spinned and inserted in duplicated into a Glucose Analyzer II (Beckman Instruments Inc. Palo Alto, CA, USA). BG dropped to a mean value

Key-words: Somatosensory evoked potentials. induced hypoglycemia. Correspondence: Gu,do Tamburrano. MO. Islituto Clinica Medica II. Policlinico Umberto I. Universita di Roma "La Sapienza". 1-00161 Rome. Italy

Received September 2, 1991.' accepted February 20. 1992.

279

G. Tamburrano, N. LOGuratolo, G. Pozzessere, et al.

of 1.90 ± 0.13 mmol/L. At the end of 90 min study, the subjects received an iv injection of a bolus of glucose to restore blood glucose concentration to approximately 6 mmol/L. Group B: six subjects (3 men and 3 women, aged 25-29 yr) in whom a moderately prolonged hypoglycemia was induced by a continuous iv infusion (0.83 mU/kg/min for 120 min). The initial blood glucose concentration decreased to about 2.4 mmol/L after 90 min. These blood glucose levels were maintained for 1 h by continuous infusion of glucose, adjusted according to BG findings. At the end of the study the insulin infusion was stopped and subjects were given iv glucose. Five more normal subjects were studied with the same study protocol of group B, but with glucose maintained at basal level (glucose clamp technique). All the subjects gave informed consent to the study which started at 09.00 h after overnight fast. In both groups SEP recordings were obtained before the induction of hypoglycemia and every 15 min throughout the experimental period, using an electric stimulus (a square wave of 0.1 msec duration and 7 Hz frequency) with surface electrode on median nerve at the wrist. 512 stimuli were emitted during each test (analysis time 50 ms), and coefficient of variation was less than 0.25% in each test. More details of this technique have been reported elsewhere (10). The temperature of the stimulated limb was maintained at 36 C by means of a thermistor-infrared lamp system. The following parameters were considered: wrist-Erb conduction velocity (CV) (representative of CV along the peripheral nerve fibers from the wrist to the brachial plexus); conduction time (CT) between brachial plexus and the cervical cord (Erb-N13) and from the cervical cord/lower brain stem lemniscal pathways to the cortex (N 13-N 20 ) (11, 12). Following the American EEG Society guidelines, the limit between normal and pathological values for electrophysiological findings was taken as mean ± 3 SD of reference group (13). Erb-N13 and N 13-N 20 CTs, and wrist-Erb CV values of 3 SD above and below the mean values of the reference group, re-

spectively, were considered pathological. Statistical significance was evaluated using preliminary twoway anlysis of variance and, when appropriate, Student's two tailed paired ttest. A probability level of p < 0.05 was considered statistically significant.

RESULTS Mean blood glucose nadir was 1.90 mmol/L in group A and 2.38 mmol/L in group B (p < 005) at 30 and 60 min, respectively. The rates of fall of blood glucose levels were 0.09 mmol/L/min during the time interval of 0-30 in group A, and 0.026 and 0.022 mmol/L/min during 0-30 and 0-60 min in group B. In both groups, basal values of the wrist-Erb CV, Erb-N13 and N 13-N 20 CTs, were all within the normal range. In group A, the electrophysiological parameters taken into consideration did not change significantly throughout the study (Table 1). In group B, the mean N 13-N 20 transit time increased significantly from a basal value of 5.82 ± 0.17 to 6.22 ± 0.19 msec at 105 min (p < 0.02) and 6.33 ± 0.25 msec at 120 min (p < 0.05), (Table 2). None of these values was in the pathological range. In the five subjects studied in euglycemia, all the electrophysiological parameters taken into consideration did not change significantly throughout the study.

DISCUSSION The present study indicates that the wrist-Erb conduction velocity (CV), which represents the CV along the peripheral nerve fibers from the wrist to the brachial plexus, is not impaired during both acute and moderately prolonged hypoglycemia in normal subjects. Recently, acute severe hypoglycemia, defined as the decrease of blood glucose levels below 2 mmol/L 2 h after a long-acting insulin injection, has been reported to induce a 36% decrease of the anterograde fast component of axonal transport in non diabetic rats (6). In that study,

Table 1 - SEP values (mean ± SE) during acute hypoglycemia in normals (group A). Time (min) BG (mmol/L)

0

15

30

45

60

90

4.21 ± 0.19

2.45±0.18

1.90± 0.13

2.75±0.15

3.12±0.11

3.62±013

wrist-Erb CV (m/sec)

71.7±3.0

70.5 ± 2.0

71.2±1.1

69.9 ± 2.1

69.8 ± 2.0

69.4 ± 2.6

Erb-N13 CT (msec)

4.40 ± 0.7

4.25 ± 0.6

4.22 ± 0.7

4.40 ± 0.2

4.45 ± 0.1

4.43 ± 0.1

N13-N20 CT (msec)

5.95 ± 0.8

6.15 ± 1.0

590 ± 1.1

6.15±0.9

6.10 ± 1.0

5.95 ± 19

BG = Blood Glucose; CV = Conduction Velocity; CT = Conduction Time. Normal Values (mean ± SO): wrist-Erb: 72.2 ± 3.6 m/sec; Erb-N13: 4.3 ± 0.4 msec; N13-N20: 6.0 ± 0.4 msec.

280

SEP in hypoglycemia

Table 2 - SEP values (mean ± SE) during sustained hypoglycemia in normals (group B). time (min)

0

15

30

45

60

75

90

105

120

BG (mmol/L)

4.22 ± 0.18 3.83 ± 0.34 3.05 ± 0.31 2.50 ± 0.25 2.38 ± 0.16 2.55 ± 0.12 244 ± 0.08 2.38 ± 0.05

245 ± 0.07

wrist-Erb CV (m/sec)

70.1 ± 0.9

69.9 ± 1.3

70.2 ± 1.2

70.0 ± 1.3

69.1 ± 14

70.1 ± 1.0

71.1 ± 1.2

70.3 ± 1.2

70.8 ± 0.9

Erb-N13 CT (msec)

4.52 ± 0 1

4.63 ± 01

4.42 ± 0.2

4.33 ± 0 1

4.37 ± 04

4.33 ± 01

440 ± 0 1

4.37 ± 0 1

4.54 ± 01

N13-N20 CT (msec)

5.82 ± 0.1

5.97 ± 0.2

5.93 ± 0.2

6.16±0.2

6.00 ± 0.1

6.09 ± 0.2

5.99 ± 0.2

6.22 ± 0.1 ** 6.33 ± 0.2*

BG = Blood Glucose; CV = Conduction Velocity; CT = Conduction Time. p < 0.05, ** P < 0.02 vs baseline. Normal Values (mean ± SO): wrist-Erb: 72.2 ± 3.6 m/sec; Erb-N13: 4.3 ± 04 msec; N13-N20: 6.0 ± 04 msec.

*

have affected the neurophysiological measurements. Notably, we and other authors have previously reported that in normal subjects a similar degree of hypoglycemia is capable of inducing electroencephalographic alterations predominantly in the frontal regions (3, 14). Moreover, cognitive function assessed by means of auditory event-related potentials has been recently reported to be impaired at higher plasma glucose concentrations (4, 15). In conclusion, our study indicates that in normal subjects blood glucose levels ranging from 1.90 mmol/L to 2.40 mmol/L do not influence the peripheral nerve function. Finally, hypoglycemia as low as 2.4 mmol/L can induce subtle impairments of the central somatosensory pathways in normal subjects only when kept constant for a time interval longer than 60 min.

the transport velocity was unaffected and the reduction in the amount of the anterograde fast component of axonal transport was prevented by pretreatment with insulin for three days, suggesting an adaptation of neuronal metabolism to hypoglycemia. In our study, no alterations in the median CV were evident during both acute and moderately prolonged hypoglycemia in normal subjects. Of note, the different nadir of plasma glucose between group A and B (1.90 mmol/L vs 2.38 mmol/L) and the different rates of falling in the glycemic levels (0.09 mmol/L. min vs 0.022 mmol/L. min), did not affect the peripheral nerve conduction velocities. The N13-N20 interval, which represents the transit time from the cervical cord/lower brain stem lemniscal pathways to the cortex, is exclusively dependent on a central somatosensory pathway. Due to the high susceptibility of the brain to the hypoglycemic injury, the finding of a sligh increase in N13-N20 CT during moderately prolonged hypoglycemia in normal subjects, is not surprising. It is worthwhile noting that similarly to other authors (9) we did not find any change of the N13-N20 interval during acute hypoglycemia. In our study, 60 min of plasma glucose levels below 2.5 mmol/L have been required for some alterations to become evident. Their clinical relevance, however, remains to be clarified, since none of the N 13N 20 intervals was pathologic, i.e. more than 3 SO above the mean values of control subjects in our laboratory. Interestingly, other authors (14) have recently demonstrated a cerebral adaptation in normal subjects after 60 min of mild hypoglycemia. Moreover, we included in the protocol a control group of five subjects in whom we infused the same amount of insulin of protocol B but without inducing hypoglycemia (glucose calmp technique). Since no statistical variation was observed in these subjects, we can exclude that hyperinsulinemia per se may

ACKNOWLEDGMENTS This research was performed within the Hypoglycemic Syndromes Program (n. 4.5 Cardiovascular and degenerative diseases and non-infectious diseases - Istituto Superiore di Sanita), and was granted in part by the Centro Internazionale Studi Oiabete (CISO) Roma. We thank E. Maroccia, O. Zarrella, G. Oi Giuseppe and G. Ricciardi for their technical assistance.

REFERENCES 1. Robinson PF, Rapoport S.I. Glucose transport and metabolism in the brain. Am. J. Physiol. 250: R127, 1986. 2. Green DA, Winegrad AI. In vivo studies of the substrates for energy production and the effects on glucose utilization in the neural components of peripheral nerve, Diabetes 28.' 878, 1979.

281

G. Tamburrano, N. LOGurat%, G. Pozzessere, et at.

3. Tamburrano G., Lala A, Locuratolo N., Leonetti F., Sbraccia P., Giaccari A, Busco S, Porcu S. Electroencephalography and visually evoked potentials during moderate hypoglycemia. J. Clin. Endocrinol. Metab. 66. 1301,1988.

9. Agarth C.D. Neurophysiological changes during insulin-Induced hypoglycemia and the recovery following glucose infusion in type 1 (insulin-dependent) diabetes mellitus and in normal man. Diabetologia 33. 319, 1990.

4. De Feo P., Gallai V., Mazzotta G, Crispino G.,

10. Pozzessere G., Rizzo PA, Valle E. Early detection of neurological involvement In 100M and NIDDM. Diabetes Care 11. 473, 1988.

Torlone E, Perriello G., Ventura MM, Santeusanio F., Brunetti P, Bolli G.B. Modest decrements in plasma glucose concentration cause early impairment in cognitive function in absence of hypoglycemia symptoms in normal man J. Clin. Invest. 82.· 436,1989.

11. Anzioska B, Cracco R.Q.

Short latency SEP's to median nerve stimulation: Comparison of recording methods and origins of components. Electroencephalogr. Clin. Neurophysiol. 52. 531. 1981.

5. Sharma AK., Duguid IGM, Blanchard OS, Thomas P.K. The effect of insulin treatment on myelinated nerve fiber maturation and integrity and on body growth in streptozotocin diabetic rats. J. Neurol. Sci. 67. 285,1985.

12. Chiappa KH, Choi SK, Young R.R. Short latency somatosensory evoked potentials following median nerve stimulation in patients with neurological lesions. In Desmedt J.E. (Ed.), Progress in clinical neurophysiology. Clinical uses of Cerebral, Brainstem and Spinal Somatosensory Evoked Potentials. Ed. Jerger, Basel 1981, vol. 7, p. 264.

6. Sidenius P., Jakobsen J. Anterogrades fast component of axonal transport during insulin-induced hypoglycemia in non-diabetics rats. Diabetes 36. 853,1987.

13. American Electroencephalografic Society. Guidelines for clinical evoked potentials studies J. Clin. Neurophysiol. 3 (Suppl. 1) 45, 1986.

7. Jaspan J.B., Wollman R.L., Bernstin L, Rubenstein AH. Hypoglycemic peripheral neuropathy in association with insulinoma: implication of glucopenia rather than hyperinsulinism. Medicine 61(1): 33,1982.

14. Kerr 0, McDonald lA, Tattersal R.B. Adaption to mild hypoglycemia in normal subjects sustained increases in counterregulatory hormones. Diabetologia 32. 249, 1989.

8. Burke D., Skuse N.F, Lethlean AK. Cutaneous and muscle afferent components of the cerebral potentials evoked by electrical stimulation of human peripheral nerves. Electroencephalogr. Clin. Neurophysiol. 51." 579, 1981.

15. Blackman J.D., Towel V.L, Lewis GF, Spire GP, Polonsky K.S. Hypoglycemic thresholds for cognitive dysfunction in humans. Diabetes 39. 828, 1990.

282