High Voltage Stimulation

High Voltage Stimulation Effects of Electrode Size on Basic Excitatory Responses GAD ALON The purpose of this study was to examine the effects of electrode size on thresholds of sensory and motor excitation, strongest motor excitation without pain perception, and strongest motor excitation coupled with maximally tolerated painful stimulation. A high voltage stimulator with monophasic pulsatile current was applied to the quadriceps femoris muscles of 14 healthy subjects. Voltage output, pulse-charge density, and isometric muscle torque were measured during random application of four electrode pairs measuring 3 x 3, 6 x 6, 9 x 9, and 5 x 16.2 cm. Results indicated a dependence of the measured variables on electrode size; the larger electrodes required greater voltage output but less pulse-charge density than the smaller electrodes. The two largest electrodes evoked significantly greater nonpainful and maximally tolerated painful muscle torques. Maximal volitional contraction increased 13.3% after completion of all stimulations. Electrode size should be considered by physical therapists when administering transcutaneous electrical nerve stimulation. Key Words: Electrodes, Motor neurons, Pain, Physical therapy, Sensory thresholds.

Basic excitatory responses to transcutaneous electrical nerve stimulation (TENS) can be divided into three general groups: sensory, motor, and painful stimulation. Excitation of large myelinated sensory A-fibers, particularly group II, which conduct predominantly pressure and tactile input,1 is usually perceived as a tingling sensation. This perception has been defined as sensory stimulation. Electrically induced motor nerve excitation causing either twitch or tetanic muscle contraction has been recognized as motor stimulation. Small myelinated A-fibers (group III) and nonmyelinated C-fibers (group IV) conduct predominantly pain and temperature information.1 If these fibers are electrically excited, the subject will perceive pain. My associates and I reported recently that a very short pulse duration provided better perceptual discrimination among the excitation of sensory, motor, and painful stimulation when surface electrodes were used.2 Furthermore, as pulse duration was shortened, less pulsecharge density (PCD) was required to elicit each of these three excitatory responses. Because electrode size also affects PCD, the size of electrodes may also contribute to such physiological disDr. Alon is Assistant Professor, Department of Physical Therapy, School of Medicine, University of Maryland, 32 S Greene St, Baltimore, MD 21201 (USA). This study was conducted at the Lewis National Prosthetic Institute, Tel-Hashomer, Israel. This article was submitted December 23, 1983; was with the author for revision 28 weeks; and was accepted November 19, 1984.

890

crimination. The effect that electrode size has on these physiological responses, however, has been almost completely overlooked in available literature. Milner and colleagues compared effects of different electrode sizes on muscle force and pain.3 One positive electrode was 77.4 cm2, and the second negative electrode was varied at 5.4, 12.9, 25.8, and 51.6 cm2. By ignoring that uneven electrode configuration imposes uneven current flow and uneven densities,4 they concluded that electrode size was not detrimental to force or pain generation. Milner and colleagues noted, however, that about half of the subjects preferred large electrodes when the gastrocnemius muscle was stimulated.3 Results concerning the quadriceps femoris muscle were not detailed. Recently, Hultman and associates evaluated muscle force production with two different sizes of surface electrodes and platinum-coated invasive electrodes.5 They ignored pain and actually anesthetized the skin under the small surface electrode to minimize painful stimulation. Under such conditions, Hultman and associates reported no difference in muscle force generation by different sizes of surface electrodes.5 Clinicians are currently using TENS for a variety of clinical problems. Some applications require only sensory stimulation6; others call for motor excitation resulting in nonfatiguing tetanic contraction without pain.7-9 Still other applications are set for painful stimula-

tion.10 Optimal treatment protocols in each of these applications may depend, in part, on electrode size. The purpose of this study was to determine the effect of electrode size on perceptual discrimination between sensory, motor, and painful responses and to measure the differences in PCD in relation to changes in electrode size. The null hypothesis was that electrode size will not affect these aforementioned excitatory responses. METHOD Subjects The subjects (eight women and six men) were informed of the purpose, procedure, benefit, and possible risks of the research after which they verbally consented to participate in the study. All were physical therapy students at the Hain-Sheba Medical Center, Israel, and had no eventful medical histories of their lower extremities. The subjects' mean age was 25.2 years (range, 21 to 32 years). Equipment I obtained force measurements with a 221A03 piezoelectric force transducer* connected to a Digital Electrometer model 615.† This force system was calibrated using 40 kg at 1-kg increments of weights. The recorded voltages were fitted by linear regression method, * PCB, Piezoelectronics, Inc, Buffalo, NY 14043. † Keithley Instruments, Cleveland, OH 44139.

PHYSICAL THERAPY

RESEARCH which later served to convert the exper imental voltage reading to force values in newtons. The high voltage stimulator used was a Model 450 Microdyne.‡ Its output included a monophasic pulse of 5 to 20 µsec duration, a peak voltage extending to 500 V, and a pulse rate set at 80 pps. Two output capacitors equal to 0.025 µF, multiplied by the output voltage, yielded a pulse charge that was 12.5 µC at maximum voltage. The stimulating electrodes were made of carbon impreg nated silicon and measured 3 x 3, 6 x 6, 9 x 9, and 5 x 16.2 cm. The respective electrode areas were 9, 36, 81, and 81 cm2. The latter two pairs were the same size, but one pair was a square and the other pair was a rectangular configura tion.

Procedure Each subject was seated and posi tioned with the test knee at 30-degree flexion and the hip at 100-degree flex ion. The thigh was secured to the table to prevent movement. I connected the ‡ Chattanooga Corporation, PO Box 4287, Chat tanooga, TN 37405.

TABLE 1 Two-way Analysis of Variance of Voltage Output TABLE 1 Source df Electrode size (A) Physiological responses (B) Interaction (AxB) Error a

3 3 9 208

MS

F

1305300.00 183323.77 12632.62 3700.44

352.74a 49.54a 3.41a

Significant at α = .01 level.

force transducer to the leg segment with a cable and adjusted the cable to form a 90-degree angle with the leg. The dis tance between the knee axis and force line was measured and used later to calculate knee-extension torque. Before the application of electrical stimulation, each subject repeated 15 trials of maximal volitional contraction (MVC), with a one-minute rest between each trial, to become familiar with the testing procedure and to eliminate the effect of learning on the test results. Later, after a five-minute rest, three trials of MVC were recorded with a oneminute rest after each trial. These MVCs served as the prestimulation maximal force values for each subject. I averaged the three trials to yield the pretest MVC score.

The electrical stimulation procedure included applying the four different pairs of surface electrodes over the quad riceps femoris muscle in a random or der. I placed one electrode of each pair at the femoral triangle overlying the femoral nerve. The second electrode was placed 10 cm superior to the base of the patella. The two electrodes constituted a bipolar technique of electrode place ment, with the proximal one negative (-) and the distal one positive (+). A stimulation test protocol was pre ceded by stimulation with a gradual in crease of current intensity so that the subjects could get familiar with the per ception and discrimination of threshold sensory, threshold motor, threshold pain, and maximally tolerated painful stimulation. The actual testing then be-

Fig. 1. Effect of voltage on basic physiological responses. Electrode sizes are A = 3 x 3 cm; B = 6 x 6 cm; C = 9 x 9cm; D = 5 x 16.2 cm. Volume 65 / Number 6, June 1985

891

TABLE 2 Newman-KeuPs Post-Hoc Test on Voltage Output Electrode Size 3 x 3 cm

Electrode Size 6 x 6 cm

Physiological Responses

Physiological Responses

Sensory

Motor

Maximal Motor

Maximal Painful

Sensory

Motor

Maximal Motor

Maximal Painful

1

2

3

4

1

2

3

4

1 2 3 4

5.11

a

a

9.57 4.45a

a

16.72 11.61a 7.16a

1 2 3 4

4.50

Electrode Size 9 x 9 cm Sensory 1 2 3 4 a

Motor

Maximal Motor

Maximal Painful

6.56a

17.51a 10.95a

23.06a 16.50a 5.54a

Sensory 1 2 3 4

12.89 8.39a

17.08a 12.58a 4.18a

Motor

Maximal Motor

Maximal Painful

9.42a

26.00a 16.58a

31.99a 22.57a 5.99a


Source

df

MS

F

Electrode size Contraction type Interaction Error

3 1 3 104

3528.96 2303.60 226.23 315.20

11.19a 7.31a NS


gan. As the intensity was increased very gradually, each subject indicated when he or she began to feel a flow of current. At this instant, the voltage, which indi cated the threshold of sensory response, was recorded. I then increased intensity until I observed a minimal, visible te tanic muscle contraction. Voltage of this motor threshold response was also re corded. The augmentation of intensity coincided with increased muscle con traction until the subject perceived pin prick pain under the electrode. The in tensity was then slightly decreased to eliminate that pain, and the stimulator voltage and the muscle force were doc umented as representing strongest mo tor stimulation without pain. I defined this intensity level as maximal motor excitation without pain. As the current was intensified further, a stronger con traction and greater pain were evi denced. When a subject indicated he or she could not tolerate any more pain, I recorded the muscle force and stimulat ing voltage again and defined this as maximal motor excitation plus maximal pain perception. This protocol was re peated with each pair of electrodes. After 892

a

Electrode Size 5 x 16.2 cm

TABLE 3 Two-way Analysis of Variance of Muscle Torque

a

a

stimulation, I recorded three more trials of MVC, constituting poststimulation maximal force values. Averaging these MVCs yielded the posttest MVC score.

Data Analysis I calculated knee extension torque Tk produced by either MVC or electrical stimulation using the equation: Tk = F x d + Tg

(1)

where F was the measured force in newtons, d was the distance between the knee axis and the location of the cable on the leg segment, and Tg was the gravitational torque that opposed knee extension. I calculated Tg using the equation: Tg = Ms COS 30° x L

(2)

where Ms was the segment mass calcu lated as a percentage of total body weight using data of Clauser and co workers,11 and L was the distance be tween the center of mass and the kneejoint axis.11 Mean leg plus foot mass was 3.83 kg ± 0.7 s. The four electrode sizes and four physiological responses were subjected

to two separate analyses of variance (ANOVAs) corrected for repeated meas ure. Dependent variables of the two re spective tests were voltage output and PCD. The latter was calculated by divid ing the pulse charge by the electrode surface area. I used an additional ANOVA to analyze the effect of electrode size on muscle-torque generation. A paired t test was used to compare the pretreatment and posttreatment torque values of MVC. RESULTS The mean voltage values required to elicit each of the physiological responses with the various electrode sizes are illus trated in Figure 1. As electrode size in creased, each response required more voltage, which resulted in the increase in the mean output voltage. Variations among subjects can be depicted from the large standard deviation. Despite the variability, I found a statistically signif icant difference (p < .01) between the voltage intensities required to reach sen sory threshold, motor threshold, strong est motor excitation without pain, and strongest motor excitation plus painful stimulation. Table 1 summarizes the ANOVA test for the four pairs of elec trodes. The Newman-Keuls post-hoc test in dicated that each of the investigated physiological responses required a sig nificantly greater (p < .01) stimulating voltage as larger electrodes were tested (Tab. 2). Table 3 summarizes the influence of electrode size on the ability to generate PHYSICAL THERAPY

RESEARCH strongest muscle torque without pain and strongest muscle torque with maxi mal toleration of pain. The ANOVA test indicated a significant difference be tween electrode size and contraction type at α = .01, but insignificant inter action. The Newman-Keuls post-hoc test (Tab. 4) revealed that both the 9- x 9- and 5- x 16.2-cm electrodes evoked a significantly greater nonpainful torque than the 3- x 3-cm electrode. In addi tion, the 5- x 16.2-cm rectangular elec trodes were statistically superior to the 6- x 6-cm electrodes but only at the α = .05 level. Very similar results of sta tistical significance between the various electrodes were present when the stim ulation of strongest muscle torque was coupled with highest level of pain toler ance. The mean magnitudes of torque achieved by using the different electrode sizes are illustrated in Figure 2. The effect of electrode size on PCD is depicted in Figure 3. In general, smaller electrodes required greater charge den sity to elicit each of the four excitatory states (Tab. 5). The Newman-Keuls post-hoc test established that an elec trode size of 3 x 3 cm required signifi cantly greater PCD than all other elec trode pairs, and the 6 x 6 cm required significantly greater PCD per unit area than the 9- x 9- and 5- x 16.2-cm electrodes. The latter two electrodes were identical in surface area, but configuratively different. No statistical dif ference in PCD could be demonstrated between them (Tab. 6). Finally, I observed that the mean MVC of the poststimulation was greater than the prestimulation values by 13.3%. A paired t test revealed the poststimulation MVC as significantly stronger at the α = .01 level (Tab. 7). DISCUSSION My colleagues and I recently reported the following sequential order of periph eral nerve excitation by TENS: 1) sen sory, 2) motor thresholds, 3) strongest motor without pain, and 4) strongest motor plus painful stimulation.2 That sequence was present when 2- x 2-cm electrode size was constant and pulse duration was varied. The present results extend such a sequence to include the conditions where pulse duration is fixed and electrode size is variable. My present results have some inter esting implications because of the des ignation of the four electrode pairs as Volume 65 / Number 6, June 1985

Fig. 2. Influence of electrode size on muscle torque generation. (A,B,C,D—refer to Fig. 1.)

independent and PCD as dependent variables. Electrotherapy literature has long recognized that the smaller the elec trode, the greater the PCD.4, 12 Figure 3 suggests, however, that this dependence is not linear in nature and that the small electrode area demands a far greater pulse charge concentration to excite the peripheral nerves. Whereas the quan tities of PCD required to account for all four excitatory levels is not directly available from my study's results, a reader may deduce from Figure 3 that a range between 0.12 to 0.25 µC/cm2 may include optimal PCD. This thinking is restricted to surface electrode areas be tween 9 and 81 cm2 as applied to the quadriceps femoris muscle. The calculated torque of the quadri ceps muscle, induced by the stimula tion, increased as a function of electrode size. The 9- x 9- and 5- x 16.2-cm electrode pairs were shown to be signif icantly advantageous over the smaller electrodes for either nonpainful or pain ful motor stimulation. These findings contradict the results of Milner and col leagues.3 Uneven electrode size, elec trode position, and absence of statistical testing in their study may be three rea sons for the disparity. The two larger pairs of electrodes were of equal area, 81 cm2, but different in shape. The rectangular pair generated

an average of 44.7 N.m, compared with 40.2 N.m, the strongest painful torque generated by the square-shaped pair. Limitation of the stimulator's electrical output may have been the reason for an absence of statistical significance. Eleven of 14 subjects stated that they could have tolerated more pain, but the output of 500 V and 12.5 µC of pulse charge were the upper limit of this clin ical high voltage generator. Further elab oration on most appropriate electrode TABLE 4 Newman-Keul's Post-Hoc Test Maximal Torque Without Pain 3x3 (cm) 1 2 3 4

Electrode Size 6x6 9x9 (cm) (cm) 1.81

4.54a 2.73

5 x 16.2 (cm) 5.51a 3.70b 0.98

Maximal Torque + Maximal Pain 3x3 (cm) 1 2 3 4 a b

6x6 (cm)

9x9 (cm)

5 x 16.2 (cm)

5.72a

8.42a 2.70

9.43a 3.71b 1.01

Significant at α = .01 level. Significant at α = .05 level.

893

TABLE 5 Two-way Analysis of Variance of Pulse Charge Density

a

Source

df

MS

F

Electrode size (A) Physiological responses (B) Interaction (A x B) Error

3 3 9 208

1.01 2.20 0.24 0.01

126.90a 277.29a 30.43a


size for the quadriceps femoris muscle must wait until a stimulator with higher voltage or greater pulse charge is tested. The electrically-induced torque of the quadriceps femoris muscle, generated by the 9- x 9- and 5- x 16.2-cm elec trodes, constituted at most only a re spective 31% and 34% of the prestimulation MVC. Nevertheless, such a stim ulation protocol led to a significantly stronger poststimulation MVC. This im provement was probably a transient re sponse because current literature sup ports a long-lasting strength gain only after several weeks of training.13-15 The immediate facilitatory effect of stimu lation may have been a posttetanic po tentiation.16 Such a facilitatory effect can be used as a preparation mode be fore a voluntary muscle strengthening program is begun.

Clinical Implications The successful eliciting of all four physiological responses under each of the electrode pairs may have some clin ical implication. A physical therapist might be able to use any short duration

pulsating current stimulator to stimulate trigger or acupuncture points by either reducing electrode size or increasing current intensity or both. This interpre tation refutes the assumption of Santiesteban that only special stimulators must be used if painful stimulation is sought.17 Should a physical therapist be using a stimulator for muscle reeducation, whereby nonpainful motor excitation is desired, electrode size should be en larged. Using larger electrodes may seem contradictory to the conclusion of Hultman and associates who reported that small and large electrodes are equally effective in excitation of maximal mus cle contraction.5 In their study, however, they anesthetized the skin under the electrodes and, therefore, blocked the painful response. Increasing electrode size first may seem reasonable clinically to optimize nonpainful muscle contrac tion. If increasing the electrode size does not produce the desired stimulation, however, the skin under the electrodes may then be anesthetized to induce greater intensity and, thus, stronger con traction.

Fig. 3. Pulse charge density as a function of electrode size. (A,B,C,D—refer to Fig. 1.)

CONCLUSION Within the scope and limitation of the present investigation, the following conclusions can be drawn: 1. Irrespective of electrode size, all basic excitatory responses (sensory, motor, pain) can be obtained by surface elec trical stimulation through the use of

TABLE 6 Newman-KeuPs Post-Hoc Test on Pulse-Charge Density Motor Threshold

Sensory Threshold 9x9 (cm)

Electrode Size 5x16.2 (cm)

...

1 2 3 4

.187

...

6x6 (cm)

3 x 3 (cm)

4.87a 4.69a

24.03a 23.85a 19.16a

...

... Motor Maximal Without Pain 5x16.2 6x6 9x9 (cm) (cm) (cm)

...

1 2 3 4

...

8.41a 7.71a

...

1 2 3 4

32.16a 31.46a 23.74a

Electrode Size 5 x 16.2 (cm)

...

.27

...

3x3 (cm)

6.65a 6.37a

38.16a 37.89a 31.51a

... Painful Maximal 5 x 16.2 (cm)

9x9 (cm) 1 2 3 4

6x6 (cm)

...

3 x 3 (cm)

... a

894

.70

9x9 (cm)

...

.35

...

6x6 (cm)

3x3 (cm)

9.35a 9.00a

46.21a 45.87a 36.87a

...

...

Significant at α = .01 level. PHYSICAL THERAPY

RESEARCH TABLE 7 Student's t Test of Maximal Voluntary Contractiona

s SE a b

Prestimulatlon

Poststimulation

t

131.61 47.02 .35

149.16 43.26 .29

3.18b

Measured in N.M. Significant at α = .01 level.

high voltage monophasic pulsed cur rent. 2. The smaller the electrode size, the lower the output voltage required to elicit each of the basic excitatory re sponses but the higher the PCD. The relationship between PCD and elec trode size does not seem to be linear. 3. The use of large electrodes can pro duce a stronger motor response with

out pain. Muscle contraction without pain results in less muscle torque than muscle contraction accom panied by painful stimulation. 4. Although the stimulation protocols may only induce one-third of the MVC, the immediate poststimula tion muscle torque can be increased significantly.

REFERENCES 1. Burke D, McKenzie RA, Skuse NF, et al: Cu taneous afferent activity in median and radial nerve fascicles: A microelectrode study. J Neu rol Neurosurg Psychiatry 38:855-864, 1975 2. Alon G, Allin J, Inbar GF: Optimization of pulse duration and pulse charge during transcuta neous electrical nerve stimulation. Australian Physiotherapy Journal 29:195-201, 1983 3. Milner M, Quanbury AO, Basmajian JV: Force, pain, and electrode size in the electrical stim ulation of leg muscles. Nature 223:645, 1969 4. Benton LA, Baker LL, Bowman BR, et al: Func tional Electrical Stimulation—A Practical Clini cal Guide, ed 2. Downey, CA, Rancho Los Amigos Rehabilitation Engineering Center, 1981, pp 33-35 5. Hultman E, Sjoholm H, Jaderholm-EK I, et al: Evaluation of methods for electrical stimulation of human skeletal muscle in situ. Pflugers Arch 398:139-141, 1983 6. Ali J, Yaffe CS, Sereete C: The effect of trans cutaneous electrical nerve stimulation on post operative pain and pulmonary function. Sur gery 89:507-512, 1981 7. Eriksson E, Haggmark T: Comparison of iso metric muscle training in the recovery after major knee ligament surgery. Am J Sports Med 7:169-171,1979 8. Gould N, Donnermeyer D, Pope M, et al: Trans cutaneous electrical nerve stimulation as a method to retard disuse atrophy. Clin Orthop 164:215-220, 1982

Volume 65 / Number 6, June 1985

9. Gould N, Donnermeyer D, Pope M, et al: Trans cutaneous muscle stimulation to retard disuse atrophy after open meniscectomy. Clin Orthop 178:190-197, 1983 10. Mao W, Ghia JN, Scott DS, et al: High versus low intensity acupuncture analgesia for treat ment of chronic pain: Effect on platelet sero tonin. Pain 8:333-342, 1980 11. Clauser CE, McConville IT, Young JW: Weight, Volume, and Center of Mass of Segments of the Human Body. Wright-Patterson Air Force Base, OH (AMRL-TR-69-70), 1969 12. Ray CD, Maurer DD: A review of neural stim ulator system components useful in pain alle viation. Med Prog Technol 2:121-126, 1974 13. Currier DP, Lehman J, Lightfoot P: Electrical stimulation in exercise of the quadriceps femoris muscle. Phys Ther 59:1508-1512, 1979 14. Currier DP, Mann R: Muscular strength devel opment by electrical stimulation in healthy in dividuals. Phys Ther 63:915-921, 1983 15. Romero JA, Sanford TL, Schroeder RV, et al: The effect of electrical stimulation of normal quadriceps on strength and girth. Med Sci Sport Exerc 14:194-197, 1982 16. Bishop B: Neural plasticity: Part 2. Postnatal maturation and function-induced plasticity. Phys Ther 62:1132-1143, 1982 17. Santiesteban AJ: The role of physical agents in the treatment of spine pain. Clin Orthop 179:24-30, 1983

895