Failure to take body height into considerations in the evaluation of nerve conduction velocities (CV) has recently been deemed unacceptable. This statement prompted the present study. Besides height, the influence of age, gender, and temperature was studied in 92 normal subjects, half of whom were females. The CV decreased 0.9 m/s per 10 years increase in age, the same in women and men aged 15 to 44 years. Mean temperature between distal and proximal ends of the nerve segment examined increased 6.1 % 0.3" C after heating followed by a CV increase of 7.0 % 0.5 m/s. The CV decreased 0.15 m/s per 100-mm increase in heights. When considering 37 individuals aged 25 to 34 years only, the CV increased 0.34 m/s per 1-m increase in height. In both instances, the changes were within the experimental error (2.3%) of the method. Key words: sural nerve conduction age sex temperature height experimental error MUSCLE 81 NERVE 15:666-671 1992

SURAL NERVE CONDUCTION PARAMETERS IN NORMAL SUBJECTS RELATED TO AGE, GENDER, TEMPERATURE, AND HEIGHT: A REAPPRAISAL WERNER T. TROJABORG, MD, AGNES MOON, MD, BlRGllTE B. ANDERSEN, MD, and NANNA S. TROJABORG, MD

This report was undertaken because of a provocative conclusion in a recent publication stating that "using normal limits for nerve conduction values that do not take height in consideration can no longer be considered an acceptable standard of pra~tice."~' An inverse relation between CV and height has been found in some nerves,2,18,30,37,41 -43 but not in the median nerve24,28,40 or radial nerve in females.") Because of the diversities in the literature concerning factors influencing nerve conduction parameters, we examined the possible relation between sural nerve CV and nerve action potential (NAP) amplitude and age, gender, temperature, and height. T h e sural nerve was chosen because the strongest relation of height and CV was found for this nerve.41

From the Laboratory of Clinical Neurophysiology, the National Hospital. "Rigshospitalet." University Hospital, Copenhagen, Denmark Address reprint requests to W.T Trojaborg. MD, Laboratory Clinical Neurophysiology, NF 3063, "Rigshospitalet," Blegdamsvej 9, DK 2100, Copenhagen, Denmark Accepted for publication August 5, 1991 CCC 0148-639x1921060666-06 $04.00 0 1992 John Wiley & Sons, Inc.

666

Sural Nerve Conduction

MATERIALS AND METHODS

Included in the study were 92 volunteers (46 of each sex, aged 15 to 44 years) without symptoms or signs of peripheral nerve involvement. Their height was measured as they entered the laboratory as was the temperature of their lower limbs. In 25 subjects, the sural nerve CV was determined on both sides before and after heating to examine the influence of temperature arid to determine the experimental error of the method. The remaining subjects had only 1 sural nerve investigated after preheating for 10 to 15 minutes. A thermostat was placed on the lateral surface of the leg mid-way between the lateral malleolus and mid-calf to control an infrared heater placed about 30 cm above the limb and covering the whole nerve segment examined. ?'he heater was automatically switched off when skin temperature reached about 37"C, corresponding to about 35°C near the nerve. 'J'."' Skin temperature was measured with a thermocouple connected to a galvanometric thermometer (Ellab Type T-E3). T h e ambient room temperature was 22 to 24°C. T h e mean of the temperature at the 2 sites was used in relation to conduction parameters. Near nerve electrodes were used for stimulation and recording at the lateral rnalleolus and

MUSCLE & NERVE

June 1992

Table 1. Sural nerve conduction parameters, height, and mean temperature between lateral malleolus and mid-calf in 92 normal subjects.

cv

Age group (years)

N

15-24 M+F (20.2 ? 0.3) 25-34 M+F (28.6 ? 0.3) 35 -44 M+F

(39.4 i 0.3) 15-44 M F

Ampl. (PV

Height (m)

TP (“C)

29

53 5 0.3

30

?

0.3

1.77 ? 0.25

36.0 ? 0.3

37

53 2 0.3

33

2

0.3

1.73 2 0.25

36.2

?

0.3

26

53 2 0.3

29

?

1.0

1.70 ? 0.01

36.1

?

0.1

46 46

53 53

28 35

?

1.9 2.3

1.82 t 0.01 1.66 5 0.01

36.1 36.1

2

0.34 0.12

-C

SEM

?

?

? ?

0.12 0.12

(28.9 _t 0.8) M

= males; F = females A// values exmessed as mean

’,(’

mid-calf, respectively. T h e optimal position of the stigmatic electrode was obtained by changing its position until a nerve action potential (NAP) could be recorded at a stimulus strength of 0.3 to 0.5 mA. A 3-mA stimulus of 0.2-ms duration, applied to the lateral malleolus, was enough to evoke a maximal response. T h e latency to the first positive peak of the NAP was used to calculate the maximal CV; the NAP amplitude was measured peak to peak. T h e distance between the stimulating and recording electrodes averaged 128 2 1 mm (mean k SEM, N = 92). RESULTS

T h e mean CV was the same in all 3 age classes, and was the same in females and males (Table 1). T h e linear regression line for CV versus age had a slope corresponding to a decrease in CV of 0.09 m/s per year.

CV Versus Age.

It was not our intention to determine the QlO for CV by stepwise warming of the leg, as has been done by other^,'"^,""-'^^^^ CV Versus Temperature.

Table 2. Temperature in “C (mean t SEM) at the lateral malleolus and mid-calf in 25 subjects before (cold limb) and after heating (wam limb). Malleolus lat.

Mid-calf

but to determine CV with and without temperature control. Mean values of temperature and CV from the left and right side were used. T h e temperature gradient along the nerve in the cold and warm limb was the same (Table 2). The mean temperature and CV before and after heating are shown in Table 3 , and the CV as a function of temperature in Figure 1. The linear regression line had a slope corresponding to an increase of 1.0 m/s per “C (Fig. la) at the normal limb temperature and of- 1.5 m/s after heating (Fig. lb). T h e scatter of CV values was less in the warm limb than in the “cold” (SD 1.8 m/s and 2.5 m/s, respectively). Figure l c shows the increase in CV in relation to the actual increase in temperature. For all 92 heated subjects the regression line corresponded to an increase of 1.2 m/s per “C (SD 2.0 m/s) . Nerve Action Potential Versus Age, Gender, and Temperature. T h e mean amplitude showed no

consistent difference in the 3 age groups exam-

Table 3. Sural nerve conduction parameters and mean temperature (Tp) between lateral malleolus and mid-calf in 25 normal subjects (mean age 29 1 years) before and after heating

Gradient

Unheated limb

Heated limb ~

Cold limb Warm limb Difference

29.0 t 0.4 35.0 c 0.3 6.0 t 0.4

Sural Nerve Conduction

30.7 & 0.3 37.0 ? 0.2 6.3 ? 0.3

1.7 ? 0.4 2.0 ? 0.3

CV (m/s) Amplitude (pV) TP (“0

45.0 38.0 29.9

?

2 ?

0.59 3.1 0.35

MUSCLE & NERVE

52.0 ? 0.44 36.0 ? 2.9 36.0 2 0.19

June 1992

667

1 - 1 i a) ' O o 0

.m

55 -

I

3 5 0 \

E

v

I

2 4540

I

I

~

8

35 26

27

28

29

31

30

1 32

33

1

34

Tp (Celsius)

45 40

jli

1

4

34

36

35

37

30

Tp (CJslua) 141

-,

I

Ll

-2

_

-4 -6

2

4

D

12-

,

- = - , m

.

=

=

.

E

m

=

*

6 Tp degree C

8

.

10h

. ;8 E

W . 8 .

I

I

8

v

I

> 6I

4I

04 2

4

0

6

10

Tp in degree C

FIGURE 1. individual values of sural nerve CV in 25 subjects as a function of temperature (Tp) (a) before and (b) after heating. In (a), CV mis = 16.01 + 0.99 x Tp; in (b), CV = 0.14 + 0.68 x Tp. (c) The increase in sural nerve CV as a function of difference in skin temperature of unheated and heated limbs. CV difference mis = 2.90 + 0.68 x Tp. The number of squares may be less than the number of individuals, because some have identical values. The full line represents the mean, the dotted lines represent the upper and lower 95% confidence limits.

ined (Table 1). The slope of the linear regression line for amplitude versus age (Fig. 2a) corresponded to a change of 0.03 FV per year (SD 14.5 pV). The mean amplitude was 35 2.3 p V in females compared with 28 2 1.9 p V in males ( P < 0.5). The difference in amplitude, as a function of the increase in temperature, revealed an increase of 0.3 p V / T (Fig. 2b). However, there was no consistency in this relation, as the amplitude fell 1 to 7 p V from low to high temperature in 17 subjects,

*

668

Sural Nerve Conduction

FIGURE 2. (a) Sural nerve action potential (NAP) amplitude as a function of age in 92 subjects. Note that the ordinate is a logarithmic scale. NAP KV = 29.82 + 0.05 x age. (b) Change in sural nerve action potential (NAP) amplitude as a function of increase in temperature (Tp) from an unheated to heated condition. NAP KV = 0.5 + 0.3 x Tp. Symbols and lines as in Figure 1.

remained the same in 5, and increased 1 to 2 p V in 3. Although the mean height in females was 160 mm less than in males, the mean CV was the same (Table 1). When the individual CV values were plotted against height, the slope of the linear regression line revealed a decrease in CV of 0.015 m/s per 10-mm increase in height (SD 2.4 m/s, Fig. 3a). By narrowing the age range to 25 to 34 years, the influence of height on CV was actually reversed: I-m increase in height corresponded to a 0.34 m/s increase in CV (Fig. 3b).

CV Versus Height.

Experimental error was determined from the difference (d) between left and right sural nerve CV both before and after heating using the equation E = d/0.8fl.45 It was 2.7% before and 2.3% after The Experimental Error of the Method.

heating, corresponding to 1.2 m/s in both circumstances.

MUSCLE & NERVE

June 1992

65

60

................... .- --.-.- _- . . . . . . . .

...

I

-- ....- . .... ......... . . . __ .

=

55

\

I

5

> u 50

mmm

II8

..I

m

I

.......................... ___._._.

......

45

l

40

-I 60

b'

I .........

.__...........

-

m

650

45

I

mmm

. m . .

-

_._---__.

.................

I

I I

m . . . . . ....._...^__-_...._.-..I -..-.-__.... ............... .?I

40 1 .I

1.6

1.6

1.7

1.0

2

2.1

W (m) C)

__-___

"$0 . . . . . . . ..- ...........................

._ ....

95%

3 55 \

age ranges from 6 to 84 years in nerves of upper and lower limbs.7*"1",28r33 When variation in CV due to a linear relation with height is removed, the age is not related to sural nerve CV.4' We examined a narrow age range to minimize its effect on other conduction parameters. In fact, the decrease with increasing age was found to be minimal (0.09 m/s per year). As for the influence of gender on CV, there is no overall agreement. Some authors claimed that CV ma be u to 6 m/s faster in females than in males,' 226,2873 findin s not confirmed by others, including our study.*' Several authors estimated a decrease in CV of 1 . 1 to 2.4 m/s per "C, varying in different nerves.4-6, 10- 12,19-24,35,36 When the temperature was changed stepwise, the change in sural nerve twice as much as CV was about 2 mls per 0C,6744 our estimate. Few studies did not confirm the effect of temperature on CV within ranges of 34 to 37.7"C,39,41,43 and some prefer to correct CV values to a fixed temperature of 35°C to avoid heating.3*20However, this might not be relevant under pathological circumstances, because fibers of different calibers react differently to temperature changes.16 We found a slightly higher average amplitude of the NAP in females than in males. A similar and often more pronounced difference has been noticed by others. 10,'3,'7338*42So far, an acceptable explanation for this has not been suggested, although improved conductance properties due to a thicker layer of subcutaneous fat in women could be the reason. However, the fact that the NAP amplitude of digital nerve fibers is greater in females than in males does not support this assumption. An increase in NAP amplitude by decreasing the temperature has been observed by some,4s14 but others, includin our group, found inconsistent changes.6,1033634'246 The discrepancy is most likely due to differences in recording techniques; with bipolar the slowing of CV results in less phase cancellation, a phenomenom which is avoided by monopolar recording. By focal cooling of the sural nerve, NAP amplitude increased 10 to 20 p.V when the temperature was lowered from 30 to 20"C, whereas the amplitude increased 5 to 12 p.V when the temperature rose from 20 to 35"C.31 Some authors put more emphasis on the influence of height than on tem er ature when establishing CV reference values.3,r0,4;

s

6 50 45

1 .I

1:7

1.8

1.9

2

2.1

(m)

FIGURE 3. (a) Sural nerve CV as a function of height (m) in 92 subjects. CV m/s = 56.0 - 1.56 x height (m). (b) CV versus height in 37 subjects aged 25 to 34 years. The equation for the regression line is CV (mls) = 52.41 + 0.34 x height (m).(c) Regression line (dashed) calculated from the CVlheightlage relation [sural NCV = 99.06 - 0.2665 x height (cm) - 0.0668 x age] proposed by Rivner et aL4' versus the mean regression line (full) through all actual measured CV values from the present study (MPS).

DISCUSSION

The influence of age on CV has been the subject of many reports. A linear decrease in MCV and SCV, or both, with increasing age, in the order of 0.5 to 1.8 m/s er 10 years has been described. 1,6,7,10,27,29,~,40,46This, however, may ply to some but not all nerves in the same individua1.7."'.4 1 Other reports deny any relation within

Sural Nerve Conduction

B

B

Disregarding the age of individual subjects, a difference of 22 m/s was found between the shortest

MUSCLE & NERVE

June 1992

669

and tallest subject, corresponding to a 2.5 mis decrease in CV per 10-cm increase in height, and taking age into consideration the decreases was even more marked: 3.2 m/s per 10 ~ m . We ~ ' found a 0.15 m/s decrease per 10-cm increase in height, a value well within the experimental error of the method (2.3% when the examined limb was heated). Moreover, when considering individuals aged 25 to 34 years, there was no significant relation between height and CV (Fig. 3b). T h e discrepancy between previous and present findings is illustrated in Figure 3c using the proposed equation41 to calculate the slope o f the regression line for CV versus height based on our values. It is difficult to explain such a divergence on account of anything but methodological differences. Without doubt, the use of needle electrodes limits some of the factors which determine the size of the experimental error. That is, they give (i) a better definition of the site of nerve stimulation and recording; (ii) less spread of stimulus current; (iii) a smaller stimulus artifact; and (iv) a better-defined reference point for latency measurement.6 Moreover, temperature control minimizes the individual temperature differences, which may be as much as 12°C when measured at the sole of the foot." Keference to structural nerve differences as the cause of lower CV in taller than in shorter individuals is rather ~peculative.~' There is, in fact, no anatomical evidence of nerve fiber tapering or chan es in internodal length along an individual axon! It is difficult to understand, as suggested by Rivner et al.," that the predicted normal values for CV, with height taken into consideration, increases the diagnostic sensitivity, considering the wide scatter of their conduction velocities.

REFERENCES 1. Behse F, Buchthal F: Normal sensory conduction velocity in the nerves of the leg in man. J Neurol Neurosurg P.sychiat?y 1971;34:404-434. 2. Berthold CH: Morphology of normal peripheral axons, in Waxman SG (ed): Phy.siology and Pathology of Axons. New York, Raven Press, 1978, p p 3-63. 3. Bjjiirkqvist S-E, Lang AH, Falck B, Vuorenniemi R: Variability in nerve conduction velocity. Using averages reduces it, warming of limbs does not. Electromyogr Clin NeurofJhysiol 1977;17:21-28. 4. Bolton CF, Carter K, Koval JJ: Temperature effects on conduction studies of normal and abnormal nerve. Mwcle Nerve 1982;5: 5. Bolton CF, Sawa GM, Carter K: T h e effects of temperature on human compund action potentials. J Neurol NeuroJurg Psychmtry 198 1;44:407-4 13.

670

Sural Nerve Conduction

6. Buchthal F, Rosenfalck A: Evoked action potentials and conduction velocity in human sensory nerves. Brain Res 1966;3:1- 122. 7. Burke D, Skuse NF, Lethlean AK: Sensory conduction of the sural nerve in polyneuropathy. J Neurol Neurosurg Psychiatry 1974;37:647-652. 8. Campbell WW, Ward LC, Swift T R : Nerve conduction velocity varies inversely with height. Muscle Nerve 1981;4:520-523. 9. Cape CA: Sensory nerve action potentials of the peroneal, sural and tibia1 nerves. A m J Phy.7 Med 1971;50:220-229. 10. Casey EB, Le Quesne PM: Digital nerve action potentials in healthy subjects, and in carpal tunnel and diabetic patients. j Nrurol Neurosurg P.sychiatly 1972;35:6 12-623. 11. Cunimins KL, Dorfman LJ: Nerve fiber conduction velocity distributions: Studies of normal and diabetic nerves. A n n Neurol 198 1;9:67 - 74. 12. De Jesus PV, Hausmanowa-Petrusewicz I, Barchi RL: T h e effect of cold on neve conduction of human slow and fast nerve fibers. Neurolucgy 1973;23:3 182- 1189. 13. DiBcnedeLLo M: Evoked sensory poteritds in peripheral neuropathy. Arch Phys Mrd Rehab 1972;53:126-131, 134. 14. DiBenedetto M : Electrodiagnostic evidence of subclinical disease states in d r u g abusers. Arch Phyly, Med Rehab 1976;57:62- 66. 15. Dioszeghy P: Needle arid surface recording electrodes in motor and sensory conduction studies. Electrornyvg-r Clin Neuruphysiol 1986;26:117- 122. 16. Douglas WW, Malcolm ,JL: 'l'he effect of localized cooling on conduci.ion in cat nerves.,) k'tiysifd 1955;130:53-71. 17. Felsenthal G: Median and ulnar muscle and sensory evoked potentials. A m J Phys Med 1978;57: 167- 182. 18. Gadia M T , Natori N , Kamos LB, Ayyar DK, Skyler JS, Sosenko J M : Influence of height on quantitative sensory, nerve conduction, and clinical indices ot diabetic peripheral neuropathy. Diabetes Cure 1987; 10:613-616. 19. Gassel MM, .l'rc?jaborg W: Clinical and electrophysiological study of the pattern of conduction times in the distribution of the sciatic nerve. ,I Neurol Neumsurg Psychiatry 1964; 27:351-357. 20. Geerlings AHC, Mechelse K: Temperature and nerve conduction velocity. Some practical problems. Ekckvviyogr Clin Neuruphysiol 1985;2 5 :253- 260. 21. Halar EM, IkLisa JA, Brozovich FV: Nerve conduction velocity: Relationship of skin, subcutaneous and intramuscular temperatures. Arch Phys Med Rehab 1980;61:199-203. 22. Halar EM, DeLisa ] A , Brozovich FV: Peroneal nerve conduction velocity: Importance of temperature correction. Arch Phys Med Rehab 1981;62:439-443. 23. Halar EM, DeLisa J A , Soine TL: Nerve conduction studies in upper extremities: Skin temperature corrections. Arch Phys Med Rehab 1983;64:412-416. 24. Henriksen JD: Conduction velocities in motor nerves in normal subjects and patients with neuromuscular disorders, Thesis, University of Minnesota Graduate School, 1956. 25. Ingram DA, Davis GR, Swash M: Motor nerve conduction velocity distributions in man: Results of a new computerbased collision technique. Electroenrephaloffr Clin Nrurnphy,s201 1987;66:235- 243. 26. Kemble F: General observations on sensory conduction in the normal adult median nerve. ElPctromyo~graphy 1967;7: 127- 140. 27. 1.a Fratta CW, Canestrari RE: Comparison of sensory and motor nerve conduction velocities as related to age. Arch Phy.5 Med Rrhab 1966;47:286-290. 28. La Fratta C W , Smith O H : A study of the relationship of motor nerve conduction velocity in the adult to age, sex, and handedness. Arch Phys Med Rrhuh 1964;45:407-412. 29. La Frata CW, Zalis AW: Age effect on sural nerve conduction velocity. Arch Phys Mt.cl Rehab 1973;54:475-477. 30. Lang AH, Forstriirn J , Bjiirkqvist S-E, Kuusela V: Statisti-

MUSCLE & NERVE

June 1992

cal variation of nerve conduction velocity. An analysis in normal subjects and uremic patienls. J Nwrol S t z 1977;K3:YZ-'L4 1. 31. Lang AII, Puusa A: Dual influence of temperature o n compound nerve action potential. NPurol Sci 198 1 ;51: 8 188. 32. Lang AH, Puusa A, Hynninen P, Kuusela V , Jantti V, Sillanpaa M: Evoluation of nerve conduction velocity in later childhood and adolescence. Mwcle Neive 1985;8:38-43. Levy R, Poole EW: Peripheral motor nerve conduction in the elderly demented and non-demented psychiatric patients. J Neurol Neurosurg Psychiatiy 1966;29:362-366. 34. Louis AA, Hotson J K : Regional cooling of hurnan nerve and slowed Na+ inactivation. Gleclroenci,phrilgr Chi N ~ u r o phy,siol 1986;63:371-375. Lowitzsch K, Hopf HC, Galland J : Changes of sensory conduction velocity and refractory periods with decreasing tissue temperature in n1an.J Neurol 1977;21(i: 173- 180. 36. Ludin HP, Beyeler F: Temperature dependence of normal sensory nerve action potentials. J Neurol 1977;2 16: 173180.

37. Mayer RF: Nerve conduction studies in man. Neurolqq 1963;13: 102 1 - 1030. 38. Moon A, Reid JC, Trojaborg W: Sural nerve conduction as a function of- age, sex, height and skin temperature. trot~ncephaliigrClin Neuruphysiol 1 Y85;6 1 :S1 1 1.

Sural Nerve Conduction

39. Niclsen V K : Sensory and motot- nerve conduction in the niedian nerve in normal sul?jects. Acla M P ~S t t i ~ t t l 197 3 ; 194 :435 - 44?1. 40. Not-ris AH, Shock NW, Wagirian 1 H : Age ch;iiiges i n the rriaxirnuni conduction velocity of motor fibers of hutii;iii ulnar nerves. J Ap@ Physiol 1953;5:58