Effect of Intermittent, Supine Cervical Traction on the

Myoelectric Activity of the Upper Trapezius Muscle in

Subjects with Neck Pain
DIANE U. JETTE,
JEFFREY E. FALKEL,
and CATHERINE TROMBLY

This study was undertaken to compare the myoelectric activity of the upper
trapezius muscle before, during, and after intermittent, supine cervical traction.
Twelve people with diagnosed disease or injury of the cervical spine served as
subjects. Electromyographic recordings were taken from the upper trapezius
muscle with bipolar surface electrodes. The subjects were treated with 20 minutes
of intermittent, cervical traction at a force of 8% of their body weight. Recordings
were taken with the subjects in the supine position before the traction, during
one pull and release phase of the 10th and 20th minutes of traction, and after
completion of the traction treatment. An analysis of variance with repeated
measures showed no significant differences in the myoelectrical activity during
the six time periods measured. The results of this study do not support the
clinical use of intermittent, supine traction to produce cervical muscle relaxation.
Key Words: Muscle, Neck, Physical therapy, Traction.

Individuals with neck pain commonly seek the care of horn cells in response to noxious stimuli.5 Many investigators,
physical therapists, and cervical traction is often a part of the who believe muscle spasm represents continuous, involuntary
treatment program. Disease or injury of the cervical structures muscle-fiber contraction, have sought to record spontaneous
is thought to result in muscle spasm of the supporting neck motor unit action potentials from areas of muscle spasm.6, 7
musculature, which may cause further pain.1 Cervical traction Simons, in a thorough review of the related literature, stated
is applied in an effort to stretch the involved cervical structures that the findings relating muscle spasm to muscle contraction
to relieve spasm and pain. Stretching a muscle through cer- were equivocal.2
vical traction has not been well-established, however, as an Because they believe that muscle spasm represents muscle
effective method of relieving muscle spasm. fiber contraction, many researchers advocate stretching a
Although most clinicians might agree that they can feel muscle to provide relaxation of that muscle.1, 3,6,8-11 Oppo-
areas of hardness and tenderness in muscles around an injury nents of traction, on the other hand, challenge the usefulness
and interpret these areas as muscle spasm, the nature and or propriety of cervical traction and cite evidence of electrical
etiology of muscle spasm remain controversial. Neither his- silence in areas of supposed muscle spasm.3,4,12 In 1952, Hunt
tological nor EMG examinations of areas of muscle spasm hypothesized an autogenic inhibitory role for the spindle.13
have provided clear evidence of its characteristics.2 Investiga- The group II afferents were thought to be activated when the
tors have hypothesized many causes for muscle spasm, in- muscle was placed in a static lengthened position. His findings
cluding the following: 1) decreased circulation to muscle fibers have been cited as rationale for the effectiveness of muscle
with ischemia and build-up of waste products,1 2) muscle stretch in relieving spasm.6 Recent literature, however, is not
tearing,3,4 3) irritation of the sinuvertebral nerve serving in- conclusive regarding the role of group II muscle spindle
jured ligaments or joint capsules,4 4) accumulation of irritat- afferents in autogenic inhibition, and they may even play a
ing by-products of inflammation,3 5) overstretch and hyper- role in autogenic excitation.14(p470) The success of traction has
sensitivity of muscle spindles,3 and 6) reflex firing of anterior also been attributed to reflex inhibition of muscle contraction
or spasm mediated by Golgi tendon organs,3, 6 but the inhi-
bition provided by Golgi tendon organs only moderates the
Ms. Jette is Assistant Professor of Physical Therapy, Simmons College, 300 excitatory response of the muscle to a stretch.14 (pl35). These
The Fenway, Boston, MA 02115 (USA).
Dr. Falkel was Assistant Professor of Physical Therapy, Sargent College of
neurological mechanisms may not, therefore, provide expla-
Allied Health Professions, Boston University, University Rd, Boston, MA, nation for inhibition of muscle spasm even if muscle fibers
when this study was done. He is now Assistant Professor of Physical Therapy, are actually contracting.
Department of Health and Human Services, Convocation Center, Ohio Uni-
versity, Athens, OH 45701. Jackson1 and DeLacerda15 have also suggested that
Ms. Trombly is Associate Professor of Occupational Therapy, Sargent Col- rhythmic, intermittent traction reduces pain by improving
lege of Allied Health Professions, Boston University, University Rd, Boston,
MA 02215.
circulation or by preventing or reducing adhesions and con-
This article was written in partial fulfillment of Ms. Jette's requirements for tractures of cervical structures. Traction may also reduce pain
the Master of Science degree, Sargent College of Allied Health Professions, by stimulating the large afferent fibers of muscles and joints
Boston University.
This article was submitted May 2, 1984; was with the authors for revision that presynaptically inhibit pain fiber transmission at the
six weeks; and was accepted March 11, 1985. spinal cord level.16 Traction may, however, be harmful. One

Volume 65 / Number 8, August 1985 1173

 opponent argues that neck pain is caused by damaged muscle physical therapy program for each participant. The subjects'
fibers and connective tissue and that these inflamed, torn diagnoses included whiplash, cervical radiculitis, cervical ar­
structures should not be further stretched.12 thritis, and neck strain. The subjects reported a duration of
Although several studies have been done to determine symptoms between one month and 14 years. The mean
whether cervical traction separates the vertebrae and what is duration was 2.9 years. The Sargent College Research Review
the best way to accomplish separation,8,10 only two studies Committee approved the project and the subjects signed an
have been published that measure myoelectric activity of the informed consent.
muscle being stretched during a traction treatment.15,17 One
study with healthy subjects measured myoelectric activity of Instrumentation
the lumbar muscles during inverted traction.17 An immediate Traction was applied with a Tru-trac* TT92 model traction
decrease from resting level EMG was observed during inver­ unit and a Dura-tech traction table arid head halter. Instru­
sion, but it was not sustained in the supine position. This mentation for EMG consisted of a Grass† model 7 polygraph
finding could be related to the tonic labyrinthine reflex, but with a 7P3 amplifier and a 7P10 cumulative integrator. The
the author did not address this point. In the other study, input impedance of the 7P3 is 44 MΩ differential and the
DeLacerda measured myoelectric activity in the upper trape­ half-amplitude frequency range is 3 to 75 Hz. Common mode
zius during intermittent, supine cervical traction in healthy rejection ratio is 1,600:1 and the signal to noise ratio is 280:14.
subjects.15 He concluded that myoelectric activity was higher Sensitivity was set at 50 µV/cm peak to peak. The half-
when the traction was pulling than when the traction pull amplitude frequency range of integrated electromyogram
released and that the larger the angle of pull, the higher the (IEMG) was 10 Hz to 40 kHz. The 7P10 Grass cumulative
myoelectric activity. Unfortunately, DeLacerda's interpreta­ integrator processed the EMG signal and displayed the IEMG
tion of his results is misleading. The article pictures an EMG as a ramp function with an automatic amplitude reset.
tracing that shows a 60-Hz interference signal during the pull The skin where the electrodes were to be placed was rubbed
phase. In addition, even if he did measure actual myoelectric with alcohol and lightly abraded. Beckman‡ miniature silver-
activity and not artifact, he did not control intervening vari­ silver chloride surface electrodes were placed 2.5 cm apart on
ables in the EMG recordings.18 A third, unpublished study the subject's most painful upper trapezius muscle halfway
measured myoelectric activity in cervical muscles during sit­ along a line between the acromion and spinous processes of
ting with cervical traction in subjects who had exercise- C7. The electrode impedance was less than 5,000 Ω. The
induced muscle pain.19 Myoelectric activity did not signifi­ ground electrode was either a plate placed on the ipsilateral
cantly change during or after traction. No studies have ex­ wrist or a third Beckman electrode placed on the ipsilateral
amined the effect of cervical traction on myoelectric activity acromion process. The placement of the ground electrode was
in subjects with neck pain from disease or injury. changed in an attempt to eliminate ECG artifact from the
The disparity in research findings creates difficulty in draw­ recording. The change, however, was not effective in elimi­
ing conclusions on the effectiveness or desirability of cervical nating the artifact. A head halter was placed on the chin and
traction for treating patients with neck pain. Cervical traction occiput in the standard way, and the subject lay supine on
can apply a stretch to the cervical muscles. Whether stretching the traction table with a low pillow under the head.
those muscles by traction causes an increase or decrease in
muscle electrical activity is unclear. Whether the effect of
Procedure
stretching on myoelectric activity is even pertinent to the
relief of muscle spasm and pain seems equally unclear. Pro­ The subject relaxed in the supine position for 5 minutes
ponents and opponents of cervical traction base their argu­ before the traction began, and the first IEMG recording was
ments on conflicting and often unsubstantiated theories of taken during the fifth minute of relaxation. A traction force
the nature and cause of neck pain and its relationship to of approximately 8% of the subject's body weight was then
muscle spasm. The principle of using traction to provide applied intermittently for a 20-minute period with a seven-
muscle relaxation requires further study. second pull and seven-second rest cycle. The angle of pull was
Assuming that one benefit of cervical traction is muscle 25 degrees. Eight percent of the body weight was chosen
relaxation, we hypothesized that myoelectric activity would because this percentage approximates the weight of the head
decrease as a result of traction. Therefore, the purpose of this and neck,20 and Harris stated that muscle relaxation can best
study was to determine if myoelectric activity of the upper be accomplished with a traction force that supports the weight
trapezius muscle significantly differed before, during, or after of the head.11 The mean traction force for all subjects was 5.9
intermittent, supine cervical traction in a group of subjects kg (13 lb) with a range of 4.5 to 9.1 kg (10 to 20 lb). Harris
with neck pain and muscle spasm. We expected that myoelec­ also recommended a 20- to 25-minute traction treatment to
tric activity might increase during the pull phases of traction achieve muscle relaxation.11 During the treatment, we took
but would be less after the treatment than before. four separate IEMG recordings: 1) a pull phase of the 10th
minute, 2) a rest phase of the 10th minute, 3) a pull phase of
METHOD the 20th minute, and 4) a rest phase of the 20th minute. After
the traction, the subject remained in the supine position for
Subjects
Twelve subjects, five men and seven women, with diag­
nosed neck pain participated in the study. The subjects ranged
in age from 15 to 54 years with a mean age of 38 years. All * Tru-Eze Manufacturing Co, Inc, PO Box 880, Temecula, CA 92390.
† Grass Instrumentation Co, 101 Old Colony Ave, Box 516, Quincy, MA
had palpable tender areas in at least one upper trapezius 02171.
muscle. Cervical traction was part of an established outpatient ‡ Beckman Instruments, Inc, 3900 River Rd, Schiller Park, IL 60176.

1174 PHYSICAL THERAPY

 RESEARCH

5 minutes, and a final IEMG recording was taken during the TABLE 1
fifth minute. Means and Standard Deviations of Normalized EMG Values
(mm/sec)

Time Period s
Data Analysis
Before traction 12.20 1.17
To compare subjects, we normalized our data by determin- Pull phase 10th min 12.94 1.33
ing the total IEMG (mm/sec) obtained during each subject's Release phase 10th min 12.78 1.18
experimental session. We then calculated the IEMG for a Pull phase 20th min 12.43 1.06
seven-second period under each of the six conditions de- Release phase 20th min 12.18 1.12
Five min after treatment 12.89 1.72
scribed above and converted each to a percentage of the
total.21 We did not ask for a maximal voluntary contraction
to normalize our data because we did not wish to intensify
our patients' neck pain. The IEMG values, expressed as
percentages, were then submitted for data analysis. We per- TABLE 2
formed a one-way analysis of variance (ANOVA) with re- Analysis of Variance Summary: Normalized EMG Values During
peated measures to determine if there was a significant differ- Six Time Periods of Treatment (mm/sec)
ence in the percentage of output during the six time periods. Source df SS MS F P
Subjects 11 12.07 1.10
Treatments 5 7.15 1.43 .09 NS
RESULTS Error 55 96.09 1.75
The means and standard deviations of normalized myoelec-
tric activity for each measured period of the traction treatment
are presented in Table 1. The summary of the one-way
ANOVA on repeated measures is displayed in Table 2. These malized EMG data for the pull phases of traction were,
results indicated that myoelectric activity of the upper trape- however, slightly higher than the means for the corresponding
zius muscle did not differ significantly before, during, or after release phases (Tab. 1).
cervical traction (p > .05). Higher traction forces may be needed to demonstrate a
change in myoelectric activity and also may be desirable.
DeLacerda used 13.6 kg (30 lb) of traction and believed that
DISCUSSION he demonstrated higher myoelectric activity during the trac-
tion pull than during the release phase.15 He stated that his
This is the first study, to our knowledge, that has measured findings supported the use of traction to reduce pain and
myoelectric activity of the neck muscles during cervical trac- suggested that the rhythmic muscle contraction and relaxation
tion in subjects with pain from disease or injury. Because produced by the traction increased muscle blood flow, which
traction is believed to produce muscle relaxation, we were reduced pain. Indeed, higher traction forces are believed
interested to see what effect it would have on myoelectric necessary to separate the cervical vertebrae and are, therefore,
activity of the muscle being stretched. The results of our study used by many clinicians. Valtonen et al, however, found that
do not support the hypothesis that intermittent, supine cer- in some cases, traction caused shortening of the cervical area,
vical traction, applied with a force believed to produce muscle and they attributed this to increased muscle contraction
relaxation,11 decreases myoelectric activity of the upper tra- caused by the traction.8 Both Farbman22 and Caldwell and
pezius muscle. Krusen23 found better overall improvement in patients with
In our study, once most of the subjects were lying in the cervical problems when they were treated conservatively with
supine position, we found electrical silence of the upper inpatient, low-poundage traction or with no traction or ma-
trapezius muscle. Electrical silence is expected in muscles at nipulation at all. Higher traction forces may cause protective
rest; yet, some studies have demonstrated resting motor unit contraction of the muscles around the neck. Replication of
action potentials in painful muscles.5-7 The IEMG values we our study using higher traction forces might provide impor-
obtained were reflective of ECG artifact rather than action tant additional information, especially if vertebral separation
potentials in the upper trapezius muscle. The results of this were measured at the same time as myoelectric activity.
study do not support the definition of muscle spasm as being Determining the correlation between pain reduction, vertebral
involuntary muscle contraction. Our subjects were in pain separation, and myoelectric activity would also be interesting.
and had firm, tender areas in the muscle recorded, but there Most of our subjects reported some pain reduction after the
was no evidence of contraction in the upper trapezius muscle traction treatment, but we made no attempt to correlate this
once the subject lay down. Because the cervical muscles were with the IEMG values.
in a state of relaxation with the subjects in the supine position, Other reasons for our inability to record changes in my-
further reduction in myoelectric activity with the application oelectric activity with traction may be related to instrumen-
of cervical traction was impossible. Therefore, muscle relax- tation or electrode placement. One might argue that sponta-
ation may not be an adequate explanation for the reduction neous myoelectric activity is limited to only a few muscle
of pain produced by cervical traction. fibers directly in the area of tenderness.7 In this study, the
Our findings also do not agree with those of DeLacerda, position of the electrodes was standardized for all subjects so
who found an increase in myoelectric activity of the cervical that the electrodes may not have been directly over the most
muscles during traction.15 In our study, the means of nor- tender area of the upper trapezius muscle. We also may not

Volume 65 / Number 8, August 1985 1175

have recorded from the most appropriate muscle. Perhaps CONCLUSIONS AND CLINICAL IMPLICATIONS
deeper cervical muscles are affected more by traction than
the upper trapezius muscle, and we were unable to measure The results of this study fail to support the use of intermit-
their electrical activity with surface electrodes. Other studies tent, cervical traction in the supine position as a means of
could be done that include recording myoelectric activity producing relaxation of the upper trapezius muscle. We also
directly over tender areas of the upper trapezius muscle or in did not find spontaneous resting myoelectric activity in our
other cervical muscles. subjects' upper trapezius muscles, even though the muscles
were painful. Those patients for whom the goal is cervical
Measuring EMG activity of the neck muscles in the sitting muscle relaxation may benefit from simply lying down as
position, a functional position for the head and neck, might much as from cervical traction. We recommend that physical
also reveal differences in muscle tension measured before and therapists carefully consider this information when they de-
after traction. We did not measure myoelectric activity in the cide whether cervical traction should be a part of a treatment
sitting position. program.

REFERENCES
1. Jackson R: The Cervical Syndrome, ed 4. Springfield, IL, Charles C 13. Hunt CC: The effect of stretch receptors from muscle on the discharge of
Thomas, Publisher, 1978 motoneurons. J Physiol (Lond) 17:359-379,1952
2. Simons DG: Muscle pain syndromes. Part II. Am J Phys Med 55:15-42, 14. Matthews PBC: Mammilian Muscle Receptors and Their Central Actions.
1976 London, England, Edward Arnold Publishers, Ltd, 1972
3. Arat A: Neck Sprains as Muscle Injury, Tension Headache and Related 15. DeLacerda FG: Effect of angle of traction pull on upper trapezius muscle
Conditions. El Paso, TX, Guynes Printing Co, 1973 activity. Journal of Orthopaedic and Sports Physical Therapy 1:205-209,
4. Grieve GP (ed): Common Vertebral Joint Problems. Edinburgh, Scotland, 1980
Churchill Livingstone Inc, 1981 16. Wall PD: The mechanisms of pain associated with cervical vertebral
5. Wyke B: Neurology of the cervical spinal joints. Physiotherapy 65:72-76, disease. In Hirsch C, Zotterman Y (eds): Cervical Pain. Proceedings of the
1979 International Symposium in Wenner-Gren Center. Oxford, England, Per-
6. deVries HA: Quantitative electromyographic investigation of the spasm gamon, 1972, pp 201-210
theory of muscle pain. Am J Phys Med 45:119-134,1966 17. Nosse LJ: Inverted spinal traction. Arch Phys Med Rehabil 59:367-370,
1978
7. Bruccini M, Duranti R, Galletti R, et al: Pain thresholds and electromyo-
18. Units, Terms and Standards in the Reporting of EMG Research. First
graphic features of periarticular muscles in patients with osteoarthritis of
Interim Report by the Ad Hoc Committee presented at the International
the knee. Pain 10:57-66,1982
Society of Electrophysiological Kinesiology Congress, Boston, MA, 1980
8. Valtonen EJ, Moller K, Wiljasalo M: Comparative radiographic study of the
19. Parker LA: An EMG Study of the Effectiveness of Intermittent Cervical
effect of intermittent and continuous traction on elongation of the cervical Traction on Tension Reduction of the Upper Trapezius and Posterior Neck
spine. Ann Med Intern 57:143-146,1968 Muscles. Thesis. Boston, MA, Boston University, 1978
9. Cyriax JH: Textbook of Orthopaedic Medicine: Diagnosis of Soft Tissue 20. Williams M, Lissner HR: Biomechanics of Human Motion. W B Saunders
Lesions, ed 7. Philadelphia, PA, W B Saunders Co (Bailliere Tindal), 1978, Co, Philadelphia, PA, 1962, p 15
vol 1, p 494 21. Perry J, Schmidt Easterday C, Antonelli DJ: Surface versus intramuscular
10. Deets D, Hands KL, Hopp SS: Cervical traction: A comparison of sitting electrodes for electromyography of superficial and deep muscles. Phys
and supine positions. Phys Ther 57:255-261,1977 Ther 61:7-15,1981
11. Harris PR: Cervical traction: Review of literature and treatment guidelines. 22. Farbman AA: Neck sprain. JAMA 223:1010-1025,1973
Phys Ther 57:910-914,1977 23. Caldwell JW, Krusen EM: Effectiveness of cervical traction in treatment of
12. Weinberger LM: Trauma or treatment? The role of intermittent traction in neck problems: Evaluation of various methods. Arch Phys Med Rehabil
the treatment of soft tissue injuries. J Trauma 16:377-382,1976 43:214-221,1962

1176 PHYSICAL THERAPY