Myelinated axons in peripheral nerves of adult beagle dogs: morphometric and electrophysiological measurements

Research in Veterinary Science /988. 45, /8/-/85

Myelinated axons in peripheral nerves of adult beagle dogs: morphometric and electrophysiological measurements O. ILLANES, R. MORRIS, G. C. SKERRITT, Department of Veterinary Anatomy, University of Liverpool, PO Box 147, Liverpool L69 3BX The myelinated fibre composition and conduction velocities were measured for the ulnar, saphenous and caudal cutaneous sural nerves of 10 healthy beagle dogs. A systematic random sampling technique was used to estimate the fibre diameter frequency distributions and densities. Conduction velocities were measured from evoked compound nerve action potentials. All nerves showed bimodal diameter frequency distributions with modes being approximately the same for each nerve (2 to 4 /-1m and 8 to 10/-lm or 10 to 12 /-1m). The variation in the average densities and in the shapes of histograms of the different nerves was slight; however, there was a wide variation for the same nerve in different individuals. The conduction velocities for the fastest conducting axons in the nerves ranged from 63 to 79 m s- I. These normal quantitative data will be of value in the assessment of pathological processes affecting peripheral nerves in the dog.

endoneurium and the distribution of myelinated axon sizes were determined for the ulnar, saphenous and caudal cutaneous sural nerves. Measurements made on evoked compound action potentials recorded from these nerves have been used to calculate conduction velocities. Materials and methods

Investigations were carried out on the saphenous and caudal cutaneous sural nerves, and the dorsal and palmar branches of the ulnar nerves of 10 adult (one to two years old) beagle dogs of both sexes. Before removal of the nerve biopsies electrophysiological measurements were made on the same nerves in the limbs of six of the dogs. The animals were deeply anaesthetised with sodium pentobarbitone given intravenously at a dose of 25 mg kg-I and supplemented as required. Body temperature was monitored by a rectal thermocouple and maintained NEUROPATHOLOGICAL changes affecting peri- at 37°C ± 1°C by means of a heating blanket with pheral nerves in dogs have many aetiologies (Duncan feedback control. In those dogs where electrophysio1980) and although nerve biopsies are not routinely logical measurements - were performed a muscle taken, the surgical procedures involved (Braund et al relaxant, gallamine triethiodide, was used at a dose of 1979) and their diagnostic value (Blakemore et al 2 mg kg ' ' intravenously to abolish muscle con1974) have been demonstrated. The use of electro- tractions induced by peripheral nerve stimulation. physiological methods to aid diagnosis of disease Tracheal intubation was carried out to allow artificial processes affecting peripheral nerves has also been respiration. In each animal two nerves were surgically evaluated (Lee and Bowen 1970, Holliday et al 1977, exposed at two levels; proximally for stimulation and Swallow and Griffiths 1977, Griffiths and Duncan distally for recording. The nerves were protected from 1978, Redding et al 1982, Takakura and Inada 1983, dehydration by immersion in warm liquid paraffin van Nes 1985, 1986,van Nes and van den Brom 1986). kept in a pool formed by the skin flaps. The length of In contrast with this, there is a lack of morphometric nerve exposed was kept to a minimum to prevent heat data concerning normal peripheral nerves in dogs. To loss which would reduce axon conduction velocity. the authors' knowledge the only studies that have . Stimulation was carried out with bipolar silver been reported are those of Braund et al (1979, 1980, chloride hook electrodes and an isolated voltage 1982a, 1982b) which deal with the fibre size distribu- stimulator. Recording was carried out with similar tion in the ulnar and common peroneal nerves. More electrodes arranged either to obtain bipolar recorddata on the numbers, sizes and densities ofaxons, ings, or by crushing the nerve over the most distal and of the conduction velocities in peripheral nerves electrode, to ,obtain monopolar recordings. Monoof healthy adult dogs, would be of value in the investi- polar recordings gave a clear separation of the first gation of neuropathological disorders affecting and second peaks of the compound nerve action peripheral nerves. potential and were used for measurements of the In the present study, quantitative estimations of latencies. Conduction velocities were calculated in myelinated fibre density per square millimetre of each case by measurement of the latency to the start 181

182

O. Illanes, R. Morris, G. C. Skerritt

FIG 1: Typical transverse sections of the nerves. (A) Caudal cutaneous sural nerve. (B) Saphenous nerve. (C) Palmar branch of the ulnar nerve. (01 Dorsal branch of the ulnar nerve. All nerves were photographed at the same magnification. Scale bar 100I'm

of a potential and the distance between the recording and stimulating electrodes. Measurements were made on compound nerve action potentials evoked by supramaximal stimulus intensities. Each value was derived from computer averages of potentials evoked by 64 stimuli applied at the rate of I Hz. Each measurement was repeated in two animals. The nerve biopsies were taken at the following sites: the saphenous nerve caudal to the caudal belly of the sartorious muscle, at the apex of the femoral triangle; the caudal cutaneous sural nerve 3 em distal to the popliteal fossa; the dorsal branch of the ulnar nerve near the division of the ulnar nerve in the middle of the antebrachium; the palmar branch of the ulnar nerve, also in the middle of the antebrachium. One centimetre lengths of nerve were excised and stretched gently on a card frame and then immersed in 2·5 per cent glutaraldehyde in 0'1 M sodium cacodylate buffer. The dogs were killed with an overdose of sodium pentobarbitone immediately after surgery. The nerves were post-fixed in 2 per cent osmium tetroxide, block stained in 2 per cent uranyl acetate and dehydrated in a graded series of ethanol and acetone, after which they were embedded in resin.

Semi-thin sections (0'4 ,..m thick) comprising the whole nerve in cross section were cut from each nerve sample, mounted on slides and stained with 0·05 per cent toluidine blue in I per cent aqueous borax solution. The sections were photographed with a photomicroscope and photomontages of each nerve were produced from the photographic prints at a final TABLE 1: Nerve fibre density (fibres mm- 2) in dogs

Dog number

Saphenous nerve

6346

Caudal cutaneous nerve

Ulnar nerve Dorsal Palmar branch branch

6394

8

6686

9 10

7152 7367

6458 7158 6071 6735 n14 6533 7087 5858 8166 7165

Mean

7033

6894

6567

6464

894

712

825

686

1 2 3 4 5 6 7

SO

8367

5308

6642 7295

8223 7049

5238 5001 0047 6903 6599 6614 7049 7290 5742

6390 5713 7373 6674 7062 6620 5691 5948 5714 7459

183

Myelinated axons in peripheral nerves of dogs

magnification of x 800. A stage micrometer with a 100 /Am scale was photographed at the same magnification to allow the final magnification of the light micrographs to be measured precisely. A systematic random sampling method, using a lattice of squares, based on that described by Mayhew and Sharma (1984), was used to estimate the nerve fibre diameter frequency distributions and density expressed as number mm- 2 • In each nerve the external fibre diameter was measured and where a nerve fibre appeared with an elliptical profile the smallest diameter was considered. Irregular profiles (sectioned through nodal or paranodal regions and Schmidt-Lantermann incisures) were excluded from diameter measurements but included in density counts.

(a)

•

1 ms

~

Results Typical cross sections of the nerves are shown in Fig I. The means of the estimated numbers of myelinated axons per square millimetre of endoneurium 3000 (b)

(a)

2000

1000

o

5 I'm 10

15

15

0

2000

2000

1000

1000

ld)

N I

E

E

~ LJ

i.i:

o

5 I'm 10

15

0

5 I'm 10

15

FIG 2: Histograms of the number of myelinated axons mm -2 in each of the nerves investigated. The mean values and standard deviations are plotted in 2 I'm intervals of diameter. (a) Saphenous nerve. lb) Caudal cutaneous sural nerve. (cl Dorsal branch of the ulnar nerve. (d) Palmar branch of the ulnar nerve

1 ms

FIG 3: Sample computer averaged compound action potentials. la) Recorded from the dorsal branch of the ulnar nerve, stimulus intensity 6 v, 0·2 ms duration, bipolar recording. (bl Recording from the saphenous nerve, stimulus intensity 8 v, 0·2 ms duration, monopolar recording. The arrows indicate the points taken as the onset of potentials and the asterisks the stimulus artifacts

are given in Table 1. The densities of myelinated fibres of different diameters are shown in the distribution histograms of Fig 2. All the nerves examined showed bimodal frequency distribution for myelinated axon diameters, with modes being approximately the same for each nerve; 2 to 4/Am for the small diameter peak, and 8 to 10/Am or 10 to 12/Am for the large diameter peak. The number of small myelinated fibres «6 /Am), expressed as a percentage of the total nerve fibre population was 64'7 per cent for the saphenous nerve, 58·9 per cent for the caudal cutaneous sural nerve, 56· 6 per cent for the dorsal branch of the ulnar nerve and 52' 2 per cent for the palmar branch of the ulnar nerve. The latter nerve contained small numbers of large fibres between 14 /Am and 16 /Am in diameter which were not detected in the saphenous and caudal cutaneous sural nerves. Compound nerve action potential recordings consisted of a large short latency peak, which sometimes showed multiple components followed by a smaller long latency peak. With bipolar recording the wave forms were also bipolar (Fig 3a). Using mono polar recording in paralysed animals the later peak was easily identified (Fig 3b). The calculated conduction velocities of axons giving rise to these two potential peaks are given in Table 2. Discussion In the present study some normal quantitative

184

O. Illanes, R. Morris, G. C. Skerritt

TABLE 2: Conduction velocities (m s-1) in selected nerves Nerve Saphenous Caudal cutaneous sural Ulnar dorsal branch Ulnar palmar branch

First peak

Second peak

rt 64 63 79

25

24 32 22

values of myelinated nerve fibres were established at fixed levels for the ulnar, saphenous and caudal cutaneous sural nerves in beagle dogs between one and two years old. BeMent and Olson (1977) suggested that single cross section fibre diameter histograms are not a good representation of the fibre diameter distribution because fibre dimensions do vary along the nerve, even for fibres that do not taper or branch in the length of interest. For peripheral myelinated fibres the diameter increases at the perinodal regions. decreases at the nodes and waves and wanes along the internodes (Sunderland 1978). This source of error was minimised by excluding from diameter measurements those fibres sectioned through nodes and paranodes. The myelinated fibre size population is known to be associated with the different functional composition of a particular nerve relative to the content of afferent and efferent fibres (Rexed and Therman 1948, Skoglund and Romero 1965, Boyd and Davey 1968). In the material used by the authors there was a preponderance of small myelinated fibres (~6 /lm), being greater for the saphenous nerve (64' 7 per cent of the total population of fibres), and smaller for the palmar branch of the ulnar nerve (52' 2 per cent). These findings are in accordance with the fact that cutaneous nerves (saphenous, caudal cutaneous sural and dorsal branch of the ulnar nerve in this study) contain larger numbers of small myelinated fibres than do mixed sensory and motor nerves (Stevens et al 1973). The variation in the mean densities and shapes of histograms of the different nerves was slight (Fig 2); however, there was a wide variation in the fibre densities for the same nerve in different individuals (Table I). These differences have also been reported by Ochoa and Mair (1969) in the normal sural nerve in man. In relation to the nerve fibre size distribution in the ulnar nerve, the authors' results cannot be compared with those of Braund et al (1979, 1980, 1982b) because the biopsy sites were different and the number and size distribution of myelinated axons are known to vary greatly at different levels as a result of nerve fibre tapering and branching (Sunderland 1978). The measurements of conduction velocity made on the ulnar nerve in this study are in reasonable agreement with those made in previous studies (Lee and

Bowen 1970, HolIiday et aI1977, Redding et a11982, Takakura and Inada 1983, van Nes 1985). The relationship between external axon diameter and conduction velocity have been extensively discussed (Hursh 1939, Rushton 1951, Ritchie 1982). Generally it is accepted that for myelinated peripheral nerve fibres above 2 /lm in diameter, the relationship between external fibre diameter and conduction velocity is linear. In the original observations of Hursh (1939), a slope of 6·0 was found for the graph relating conduction velocity and fibre diameter in the cat. Applying a similar calculation to the data presented here gives conversion factors for fibre diameter to conduction velocities which are lower. To clarify whether this is caused by species peculiarity or by methodological factors would require a more detailed study of individual axons in the dog. Acknowledgements We wish to record our gratitude to Professor A. S. King for his advice and encouragement, Mrs J. Henry for her technical assistance, and Mrs M. M. Thompson for typing the manuscript. References BEMENT, S. L. & OLSON, W. H. (1977) Journal ofExperimental Neurology 57, 828-848 BLAKEMORE, W. F.• MITTEN. R. w.. PALMER, A. C. & PATTERSON, R. C. (1974) Veterinary Record 94. 70-71 BOYD, I. A. & DAVEY, M. R. (1968) Composition of Peripheral Nerves. London, E. & S. Livingstone. pp 30-33 BRAUND, K. G., LUTTGEN, P. J., REDDING, R. w. & RUMPH, P. F. (1980) Veterinary Pathology 17,422-435 BRAUND, K. G., McGUIRE, J. A. & LINCONLN, C. E. (l982a) Veterinary Pathology 19,365-378 BRAUND, K. G., McGUIRE. J. A. & LINCONLN, C. E. (1982b) Veterinary Pathology 19, 379-398 BRAUND, K. G., WALKER. T. L. & VANQFVELDE, M. (1979) American Journal of Veterinary Research 40, 1025-1030 DUNCAN, I. D. (1980) Veterinary Clinics of North America, Small Animal Practice 10, 177-211 GRIFFITHS. I. R. & DUNCAN, I. D. (1978) Journal of Small Animal Practice 19, 329-340 HOLLIDAY, T. A., EALAND, B. G. & WELDON, N. E. (1977) American Journal of Veterinary Research 38,1543-1551 HURSH. J. B. (1939) American Journal ofPhysiology 127, 131-139 LEE, A. F. & BOWEN, J. M. (1970) American Journal of Veterinary Research 31.1361-1366 MAYHEW, T. M. & SHARMA, A. K. (1984) Journal of Anatomy 139,45-58 NES VAN. J. J. (1985) American Journal of Veterinary Research 46, 1155-1161 . NES VAN, J. J. (1986) Research in Veterinary Science 40,144-147 NES VAN. J. J. & VAN DEN BROM. W. E. (1986) Research in Veterinary Science 40. 189-196 OCHOA, J. & MAIR, W. G. P. (1969) Acta Neuropathologica 13, 197-216 REDDING, R. W.• INGRAM, J. T. & COLTER. S. B. (1982) American Journal of Veterinary Research 43, 517-521 REXED, B. & THERMAN, P.-O. (1948) Journal of Neurophysiology II, 133-139

Myelinated axons in peripheral nerves of dogs RITCHIE, J. M. (1982) Proceedings of the Royal Society (series B) 217,29-35 RUSHTON, W. A. H. (1951) Journal oj Physiology 115,101-122 SKOGLUND, S. & ROMERO, C. (1965) Acta Physiologica Scandinavica, Supplement 260, 66, 3-50 STEVENS, J. c., LOFGREN, E. P. & DYCK, P. J. (1973) Brain Research 52, 37-59 SUNDERLAND, S. (1978) Nerves and Nerve Injuries. 2nd edn. Edinburgh, Churchill Livingstone. pp 15-19-31

185

SWALLOW, J. S. & GRIFFITHS, I. R. (1977) Research in Veterinary Science 23, 29-32 TAKAKURA, Y. & INADA, S. (1983) Japanese Journal of Veterinary Science 45,413-416

Received March 16, 1987 Accepted October 10, 1987