Cervical Traction Device Study: A Basic Evaluation of
Home-Use Supine Cervical Traction Devices
JNMS:Journal of the Neuromusculoskeletal System
Vol.3, No2
1067-8239
Patrick P Venditti, D.C.,* Logan College of Chiropractic, Chesterfield, Missouri,
Anthony L. Rosner, PhD.,* Foundation for Chiropractic Education and Research
Arlington, Virginia, Norman Kettner, D.C., DACBR,* and Gary Sanders, PhD.,
Logan College of Chiropractic, Chesterfield, Missouri
THE DIFFERENCE IN the mechanical separation exerted upon the cervical spine by five home-use supine cervical traction devices was studied. In addition, surface EMGs recording the activity of the posterior cervical and masseter musculature were taken on five subjects during use of each of the five devices. Subject surveys were performed with 30 additional subjects after each had been placed in each of the traction devices for a period of 10 minutes. This was performed to determine how the devices were perceived by users in terms of convenience, ease of instruction, setup, and comfort.
Two supine lateral cervical views were taken of each of five subjects during testing of each of the five devices. One view was done in a neutral position, the second view was done while the subject was in the respective device after a 10-minute traction period. Radiographs were then digitized into a IBM 386 computer and analyzed using a special software program called "Spinal-HealthData" from Health Data Development. A Verimed Myo III computerized, hand-held, dual-channel, surface recording EMG instrument was utilized to record the muscle activity during traction.
The authors conclude that all the devices tested were capable of distracting the cervical spine, attaining greater separation at the posterior disc height than the anterior. From the results obtained, it was not possible to draw conclusive evidence that the traction devices have the ability to relax the muscles. Their primary feature appears to be mechanical and that is to separate the joint surfaces. Based upon the survey results, it appears that the Pronex™ device is the clear choice among the subjects tested. It scored favorably above all the other devices. However, the
C-Tract by Granberg, the Cervitrak by Staodyn, and the Necktrac by Lossing Orthopedic were rated fairly close to each other. The Pettibon device was clearly the least favored of the devices among the subjects tested.
(JNMS:Journal of the Neuromusculoskeletal System 3:82-91,1995)
For decades, cervical traction has been applied widely
for pain relief of neck muscle spasm or nerve root com-
pression (1-8). Some authors believe that traction, espe-
cially with a slight degree of neck flexion, could open the
posterior articulations, widen the intervertebral foramen,
disengage the facet surface, and elongate the posterior mus-
cular tissues and ligaments (2,4,6,8). It is a technique in
which a force is applied to a part of the body to stretch soft
tissues, to separate joint surfaces or bony structures. Betge
and So (9) demonstrated from their study that traction has
a definite effect on the cervical spine. Traction separates the
vertebral motor units, anteflexes the segmental alignment of
vertebral segments, and decreases the physiological lordotic
curvature. Cyriax ( 10) states that traction must be constant
so that the muscles may tire and the strain fall on the joints.
He further states that it takes 2 minutes of sustained traction
before the intervertebral spaces begin to widen.
Although there has been a wealth of literature supporting
the use of cervical traction, there have been no descriptive
studies of the different traction products on the market.
Many of these products are sold to health-care practitioners
to be re-sold to patients for home use. The purpose of this
study was to provide answers to three basic questions: First,
do each of the products indeed provide separation of the
vertebral segments? Second, do the devices allow for relax-
ation of the muscles around the temporomandibular joint
and the posterior cervical region? And, last, how are the
devices perceived by users in terms of convenience, ease of
instruction, setup, and comfort? The study did not attempt
to determine the clinical efficacy of the devices, but rather
dealt with relatively simple and basic issues.
MATERIALS AND METHODS
To provide answers to the first question, the initial study
involved five normal adults (three men, two women) with-
out a history of neck trauma or pain. Their age ranged from
29 to 48 years (average 39 years); body height from 167
to 187 cm (average 174.7 cm); and body weight from 54
to 102 kg (average 73 kg). The Logan College of Chiro-
practic Internal Research Review Board approved the pro-
ject and the subjects signed an informed consent.
fig. 1 Cervitrak by Staodyn
fig. 2 Pronex™ by RS Medical®
A radiographic study of the lateral cervical spine done
in a supine position with no neck traction was done on each
subject. The neck was maintained in a neutral position as
determined by goniometric measurement through the ver-
tical axis of the ear as described in the AMA Guides to the
Evaluation of Permanent Impairment, 3rd edition. The dis-
tances between radiograph machine, table, and X-ray film
were fixed. Then each subject was placed in one of five
cervical traction units: Cervitrak by Staodyn (Fig. 1 ); Pro-
nex by RS Medical®, Inc. (Fig. 2); Necktrac model #691
by Lossing Orthopedic (Fig. 3); C-Tract by Granberg In-
ternational (Fig. 4); and a simple water bag/halter traction
by Pettibone (Fig. 5). Traction angles could not be totally
controlled so that each device applied traction at the same
angle. The difference in the devices prohibited this. Trac-
tion angles varied from extension of less than 15o to flexion
no greater than 30o.
fig. 3 Necktrac by Lossing Orthopedic
fig. 4 C-Tract by Granberg
Each subject was placed into the traction device accord-
ing to manufacturer's instructions. At the end of the 10
minutes, another lateral supine cervical X-ray view was
taken. Each subject repeated this routine over the course of
3 weeks until all five traction devices were tested. The av-
erage length of time between testing for each unit was 3
days. Each subject was asked to comment about the device
after each test trial. Twenty-five test trials were completed
for a total of 50 radiographic studies. The order in which
the devices were tested were randomly chosen for each sub-
ject.
fig. 5 Pettibon traction device
Each radiograph was then digitized into an IBM 386
computer using a special software program called "Spinal-
HealthData" from Health Data Development. In order to
determine and compare joint distractions, segmental align-
ments, and changes of cervical curvatures, both before and
after traction, reference points were needed to guarantee the
greatest reliability. These points were then digitized into the
computer which then made the mathematical calculations.
The points utilized was a reference line formed by joining
two easily identifiable points on the inferior border of the
vertebrae and the addition of two points on the superior
border of the vertebrae. A comparison analysis was done
between the neutral views and the view from each re-
spective traction device. Fifty analyses were done. The an-
terior and posterior disc heights from C2 to C7 were mea-
sured by the computer. Four separate comparisons were
done. An ANOVA statistical analysis was done to compare:
1. the total anterior disc height changes for all devices (Ta-
ble 1) 2. the anterior disc heights differences at each ver-
tebral level for all devices (Tables 2-4); 3. the total
posterior disc height change for all devices (Table 5); and
4. the posterior disc height differences at each vertebral
level for all devices (Tables 6, 7).
TABLE 1. Factor ANOVA-Repeated Measures for Anterior
Disc Height Changes
Mean Scheffe Fisher
Comparison difference F test Dunnett t PLSD
Pronex™ vs. Cervitrak -4.66 1.583 2.516 3.927
Pronex™ vs. Pettibon -3.94 1.131 2.127 3.927
Pronex™ vs. C-Tract -5.76 2.418 3.11 3.927
'Significant at 95%.
TABLE 2. One Factor ANOVA-Repeated Measures for
Anterior Disc Height Changes at C2-C3
Mean Scheffe Fisher
Comparison difference F test Dunnett t PLSD
Pronex™ vs. Necktrac -0.98 2.012 2.837 0.732
Cervitrak vs. Necktrac -0.96 1.931 2.779 0.732
'Significant at 95%.
TABLE 3. One Factor ANOVA-Repeated Measures for
Anterior Disc Height Changes at C3-C4
Mean Scheffe Fisher
Comparison difference F test Dunnett t PLSD
Pronex™ vs. Pettibon -1.14 1.371 2.341 1.032
Pronex™ vs. Necktrac -1.12 1.323 2.3 1.032
'Significant at 95%.
TABLE 4. One Factor ANOVA-Repeated Measures for
Anterior Disc Height Changes at C6-C7
Mean Scheffe Fisher
Comparison difference F test Dunnett t PLSD
Pronex™ vs. C-Tract -3.54 1.289 2.271 3.305
'Significant at 95%.
TABLE 5. One Factor ANOVA-Repeated Measures for
Posterior Disc Height Changes
Mean Scheffe Fisher
Comparison difference F test Dunnett t PLSD
Pronex™ vs. C-Tract -4.62 1.31 2.289 4.278
'Significant at 95%.
TABLE 6. One Factor ANOVA-Repeated Measures for
Posterior Disc Height Changes at C4-C5
Mean Scheffe Fisher
Comparison difference F test Dunnett t PLSD
Cervitrak vs. Necktrac 0.8 1.178 2.171 0.781
'Significant at 95%.
TABLE 7. One Factor ANOVA-Repeated Measures for
Posterior Disc Height Changes at C6-C7
Mean Scheffe Fisher
Comparison difference F test Dunnett t PLSD
Pronex™ vs. Pettibon -2.78 1.51 2.458 2.398
Pronex™ vs. C-Tract -2.52 1.241 2.228 2.398
Cervitrak vs. Pettibon-2.44 1.163 2.157 2.398
Pettibon vs. Necktrac 3.4 2.259 3.006 2.398
C-Tract vs. Necktrac 3.14 1.927 2.776 2.398
'Significant at 95%.
A second study was done using the Verimed myoexer-
ciser III feedback and surface electromyogram system. In-
strumentation for SEMG consisted of a Verimed Myo III,
a computerized, hand-held, dual-channel, surface recording
EMG instrument. The instrument was designed for EMG
feedback and electromyograms. The bandwidth was 100 to
1000 Hz. The Common Mode Rejection Ratio was 100 dB.
The signal-to-noise ratio was not available from the instru-
ment. The skin where the electrodes were to be placed was
rubbed with alcohol and lightly abraded. Medtronic, Inc.
"SnapEase" silver-silver chloride surface electrodes were
placed on the masseter muscle between the zygomatic arch
and the angle of the mandible and the posterior cervical
musculature lateral to the C4 spinous process. The input
impedance was greater than 100 MSL The ground electrode
was placed on the ipsilateral acromion process. Sensitivity
was set at 40 I.x,V. Each of five subjects was then placed in
each of the different traction devices. The Myoelectric III
was placed in the CAMA (concurrent assessment of motor
activity) mode for a 10-minute interval. The CAMA mode
is for a specifically defined timed interval in order to av-
erage the integrated electromyographic signal. The aver-
aging of an integrated EMG signal is always zero or close
to it. The process of averaging the signals as defined by the
protocols for this particular device is different than one
would expect. As the device is operating, it calculates the
EMG signal at pre programmed time intervals. These figures
are stored as a set numerical signal value. At the end of the
session, these stored numerical data are mathematically av-
eraged to yield one "averaged" signal for the session. An
ANOVA statistical analysis was done to compare the results
among the various devices.
When home traction is indicated as a form of treatment,
one of the most important considerations is to prescribe a
unit that is both effective and simple to set up and use, for
this helps ensure that the patient will continue at-home
treatment on a more consistent basis ( 11 ).
A study into the effectiveness of these home-use devices
would not have been complete if we did not ask the subjects
involved about their feelings, comments, and opinions. The
most effective and best designed device would not be ben-
eficial as a treatment modality if the individual found it
cumbersome and uncomfortable to use or for some reason
rejected using the device altogether.
Thirty (30) subjects were utilized in this phase of the
study. Each subject was randomly given a cervical traction
unit. The subject read the instructions per manufacturer and
then proceeded to assemble and begin using the device. The
time it took them to set up the device and begin using it
was timed by one of the investigators. Each subject re-
mained in the device for 10 minutes. Immediately after their
test of the device, they were given an 1 I item survey (see
Appendix). Each item made a statement about the device
and the subject was to respond on a 5-point Likert scale
( I 2) from "strongly disagree" to "strongly agree." A sec-
tion for additional comments was included at the end of the
survey. This procedure was followed again for the testing
of each device until all 30 subjects had tested each of the
five traction devices. The amount of time between the test-
ing of each device averaged 3 days.
The numerical values for each question were totaled and
the arithmetic mean for each device was determined. This
number was subtracted from l0 to permit a linear compar-
ison between the different devices.
RESULTS
Radiographs of Disc Height Changes
We found that each traction device could not be set at
the same poundage so that the traction could be consistent
throughout a11 the tests. However, this situation would also
be encountered by the practitioner and patient alike. Opin-
ions differ regarding the amount of traction force necessary
to produce the desired response in the cervical spine. Some
investigators report the use of more than 90 kg (200
pounds), while others maintain that less than 4.5 kg ( 10
pounds) is adequate (9,13).
The Pettibon device utilized a water bag that, when filled
to certain marked areas, could provide as much as 20
pounds of traction. The Staodyn (Cervitrak) was preset at
18 pounds of traction; the Lossing device (Necktrac) util-
ized a patient-controlled spring device which was pulled by
the patient to the desired level, at which point the rope
would be locked into position at the top of the traction
device. A level of 30 pounds was selected for these trials.
The Granberg device (C-Tract) used a rope-and-winch sys-
tem with a preset dial system. Once the subject was posi-
tioned in the harness, he or she pulled down on the lever
arm until the traction poundage met the preset limit. We set
the limit at 55 pounds. The Pronex™ was a rubber cradle
separated by a rubber bellows. A hand-held bulb was used
to pump air into the bellows. As the bellows expanded, the
ends of the cradle, one at the base of the neck and one at
the base of the occiput, were separated. This then applied
traction to the spine. There was no way to determine or
measure the amount of force being applied. The subjects
were told to increase the air pressure to their maximum
comfort level. The lack of consistency in the amount of
traction applied did seem to have an effect on the amount
of vertebral displacement. The total mean anterior disc
height change ranged from 4.9 mm for the Pronex™ device
to 10.7 mm for the C-Tract by Granberg (see Table 8 and
Fig. 6). The total mean posterior disc height change ranged
from 8.4 mm for the Pronex™ device to 13 mm for the C-
tract by Granberg (see Table 9 and Fig. 7). The greatest
change in anterior as well as posterior intervertebral space
came from CS-C6 and C6-C7. The smallest change came
from C2-C3 and C4-C5 for both anterior and posterior
displacement (see Tables 1O and 11 ).
TABLE 8.
Total Anterior Disc Height changes for
Each Traction Device
Subject Pronex™ Cervitrak Pettibon C-Tract Necktrac
A 5.3 11.5 11.3 18.8 6.7
B 5.9 10.8 10.3 8.3 6.1
C 4.7 5.3 3.3 9.3 6.9
D 3.1 12.4 11.9 8.5 11.5
E 5.7 8.0 7.6 8.6 11.5
There was a statistical significance of the anterior disc
height differences between the Pronex™ device and the Cer-
vitrak ( 18#), and between the Pettibon ( 16-20#) and the C-
Tract (55#). There was no statistically significant difference
between the other devices. Also, there was no statistically
significant difference between the Pronex™ and the Necktrac
(30#) (see Table 1 ). There was a statistical significant dif-
ference in the total posterior disc height changes between
the Pronex™ and the C-Tract (55#) device. All others were
not found to be different (see Table 5).
The difference in anterior disc height changes at each
vertebral level among devices was generally insignificant.
There was some difference between the Pronex™ and Neck-
trac (30#) at the C2-C3 and C3-C4 levels (see Tables 2
and 3), the Pronex™ and C-Tract (55#) at the C6-C7 level
(Table 4), and the Cervitrak ( 18#) and the Necktrac (30#)
at the C2-C3 level (see Table 2).
There was, however, some statistical significant differ-
ence in posterior disc height changes specifically at the C6-
C7 level. This occurred with the Pronex™ vs. the Pettibon
( 16-20#), the Pronex™ vs. C-Tract (55#), the Cervitrak ( 18#)
vs. the Pettibon ( 16-20#), the Pettibon ( 16-20#) vs. the
Necktrac (30#) and the C-Tract (55#) vs. the Necktrac (30#)
(see Table 7). At the C4-C5 level there was a statistical
significant difference between the Cervitrak ( 18#) and the
Necktrac (30#) (see Table 6).
fig. 6 Mean anterior disc height changes (in mm)
The differences between disc height changes could not
be determined on the basis of traction poundage alone, al-
though the mean total anterior disc height changes were
greatest for the C-Tract (55#) at 10.7 mm and least for the
Pronex™ at 4.9 mm (see Fig. 6). This pattern also proved to
exist for the mean of the total posterior disc height changes
(see Fig. 7). The traction device with the most poundage
yielded the greatest anterior disc height separation; as the
tractive force decreased, so did the anterior disc separation.
fig. 7 Mean posterior disc height changes (in mm)
SEMG Results
The most significant difference here was that the Pettibon
device, which is simply a variation of the water bag and
harness traction system that has been around for decades,
caused marked muscle contraction in the masseter and the
posterior cervical musculature. There was a strong statisti-
cal difference between the Pettibon and all the other devices
(see Tables 12 and 13).
Despite the lack of muscle relaxation evident in the use
of the Pettibon device, it still was able to distract the spine.
The most common complaint heard from all subjects was
that the device was uncomfortable and the weight of the
water bag ( 16-20#) caused the harness to put extreme pres-
sure on the jaw and forehead. The muscle contractions were
most likely an attempt to resist this or perhaps compression
of the temporomandibular joint caused a reflex cervical
musculature spasm.
Subject Survey Results
No matter how clinically effective a therapy is found to
be, the treatment process, especially when it is dependent
upon home use, is highly dependent upon patient compli-
ance. The survey was designed to elicit subject responses
regarding five key areas: 1 ) the ease of the instructions for
assembly and implementation by the user; 2) ability of the
user, from a reading of the manufacturer's instructions, to
understand how the device is used and what were the con-
traindications; 3) how easy was the device to assemble; 4)
how comfortable was the device to use; and 5) what was
the perception of the user toward the effectiveness of the
device. Thirty subjects participated in the setup and use of
each device and then filled out a survey immediately after
each session.
The Pronex™ was the quickest to set up with an average
time of 3 minutes. The C-Tract by Granberg took the long-
est with an average time of 9 minutes and 55 seconds (see
Setup, Fig. 8). The survey demonstrated a clear preference
for the Pronex™ device (see Appendix).
TABLE 9.
Total Posterior Disc Height Changes for
Each Traction Device
Subject Pronex™ Cervitrak Pettibon C-Tract Necktrac
A 5.7 10.7 13.1 14.2 5.4
B 11.7 9.6 8.1 20.0 13.2
C 8.5 12.9 8.6 9.9 7.9
D 8.6 12.2 13.0 9.1 8.9
E 7.5 6.6 6.1 11.9 12.4
Out of the 11 statements on the survey, the Pronex™ scored
the best in all 11. The Granberg scored the lowest in four
categories that pertained to setup time and ease of setup.
The Pettibon scored the lowest in the remaining seven state-
ments which dealt with instructions, contraindications,
comfort, safety, and efficacy. The other devices were per-
ceived on all categories fairly closely. The Granberg was
also rated close to the Staodyn and Lossing units in terms
of safety, contraindications, instructions, comfort, and effi-
cacy.
SUMMARY
We set out to determine whether some of the home-use
cervical traction devices were capable of physically sepa-
rating joint surfaces. The results indicate that all of the de-
vices tested did separate the intevertebral joints both ante-
riorly and posteriorly. The ANOVA suggested that there
was a statistically significant difference between the amount
of disc height change with some of the devices, particularly
the Pronex™ which provided the least amount of disc height
changes (4.9 mm total anterior and 8.4 mm total posterior).
In a search of Medline for the past 40 years, we found
nothing in the literature that suggested an ideal or minimum
separation value before clinical effectiveness was demon-
strated. Therefore, we were able to conclude that all the
devices tested were capable of distracting the cervical spine,
attaining greater separation at the posterior disc height than
the anterior.
The ability for the devices to allow muscle relaxation was
a more difficult question to answer conclusively. The Pet-
tibon water bag/harness device was the only device that the
SEMG recorded significant muscular activity. Even in the
case of the Pettibon, the distraction process was not signif-
icantly impaired. Our results do not provide us with any
substantive evidence except to indicate that the traction de-
vices do not have, as a primary feature, the ability to relax
the muscles. Its fundamental feature appears to be mechan-
ical, and that is to separate the joint surfaces. The use of a
SEMG is not particularly well suited to determine ade-
quately whether muscle relaxation did or did not occur,
since it lacks the ability to receive signals from the deeper
muscle groups. A study by Jette et al. also used SEMG and
could not find significant myoelectric activities during su-
pine cervical traction ( 14).
The last question posed in the study dealt with issues
pertinent to patient compliance. Based upon the survey re-
sults, it appears that the Pronex™ device is the clear choice
among the subjects tested. It scored favorably above all the
other devices. However, the C-Tract by Granberg, and Cer-
vitrak by Staodyn, and the Necktrac by Lossing Orthopedic
were rated fairly close to each other. The Pettibon device
was clearly the least favored of the devices.
DISCUSSION
The devices tested were received through the auspices of
the Foundation for Chiropractic Education and Research
(FCER). It was their intention to test al1 the home-use su-
pine cervical traction devices on the market. Not all com-
panies volunteered their products. Of the five companies
which did provide products for testing, it was felt that the
Pettibon device was in a separate class by itself. It was the
least expensive of the products ($25-$30) and was clearly
an adaptation of the over-the-door water bag/harness device
that has been around for decades. Even though it was ca-
pable of producing adequate traction to separate the inter-
vertebral spaces, the subjects felt that it was uncomfortable,
and in many cases, painful.
The other devices ranged in price from $450 to $495 for
the Pronex™, $529 to $639 for the Necktrac by Lossing Or-
thopedic, $660 for the Cervitrak by Staodyn, and $890 for
the C-Tract by Granberg International. The devices within
these price ranges did more in terms of providing adequate
instructions and took into account the need for comfort.
The health-care practitioner who finds the necessity of
prescribing a cervical traction device for his or her patients
should be concerned with the ability of the patient to use
TABLE 10. Difference in Anterior Intervertebral Space (in mm)
After Cervical Traction in Different Traction Devices
Spine
Segment Subject Pronex™ Cervitrak Pettibon C-Tract Necktrac
C2-C3 A 0.1 1.1 0.9 0.9 2.8
B 0.8 0.2 0.6 1.7 1.0
C 0.4 0.3 0.1 0.4 1.1
D 0.8 1.5 0.7 0.5 1.3
E 1.1 0.2 1.5 1.4 1.9
C3-C4 A 0.4 0.7 2.0 0.9 0.3
B 0.7 1.1 1.1 0.8 0.6
C 0.5 0.8 1.3 1.4 2.3
D 0.2 0.4 2.8 0 1.4
E 0.6 1.3 0.9 2.0 3.4
C4-C5 A 1.3 1.7 1.5 1.2 1.4
B 0.4 0.6 1.0 0 0.3
C 0.1 1.2 0.7 0.9 0.8
D 0.9 0 0.7 1.5 2.1
E 1.3 1.7 1.1 2.1 1.3
C5-C6 A 1.2 0.3 0.6 0.6 0.6
B 0 1.3 1.1 0.2 1.3
C 1.5 0.7 0.1 3.1 0.8
D 0.3 0.9 1.8 1.5 1.3
E 0.1 1.0 0.9 0.5 1.7
C6-C7 A 0.9 0.8 4.9 11.4 0.3
B 0.7 5.6 6.3 1.0 2.3
C 0.9 0.6 0.1 3.3 1.4
D 0.5 4.1 4.2 4.6 0.3
E 0 1.2 2.2 0.4 1.3
TABLE 11. Difference in Posterior Intervertebral Space (in mm)
After Cervical Traction in Different Traction Devices
Spine
Segment Subject Pronex™ Cervitrak Pettibon C-Tract Necktrac
C2-C3 A 0.7 1.2 0.1 0.3 2.0
B 0.1 2.8 0.2 1.1 0.4
C 2.8 1.7 1.7 2.5 4.1
D 0.8 0.1 1.0 0.7 1.8
E 3.1 0.8 1.0 4.2 0.5
C3-C4 A 1.1 1.1 0.2 0.1 0.3
B 0.8 1.2 2.2 2.6 1.9
C 1.0 0.1 1.3 1.1 1.1
D 1.0 1.9 2.3 1.1 0.9
E 0.7 2.2 1.2 1.4 6.3
C4-C5 A 0.8 0.1 0.4 1.5 0
B 1.5 1.0 1.0 0.6 0.7
C 0.7 2.1 0.1 0.9 0.5
D 1.9 1.9 2.0 1.3 0
E 0.3 0.9 0.7 1.0 0.8
C5-C6 A 0.7 1.7 2.0 1.5 0.1
B 1.6 1.9 1.4 4.7 2.7
C 0.9 4.6 0.9 1.7 1.4
D 1.6 0 0.8 1.7 0.6
E 1.8 2.5 0.6 0.4 0.8
C6-C7 A 2.2 2.3 9.8 9.7 2.0
B 2.8 2.3 3.2 3.9 0.7
C 0.5 1.8 1.0 3.5 0.5
D 0.9 1.9 4.7 0.2 0.3
E 0.4 0.2 2.0 2.1 0.2
TABLE 12. One Factor ANOVA-Repeated Measures for
EMG Readings at Masseter Muscle
Mean Scheffe Fisher
Comparison difference F test Dunnett t PLSD
Pettibon vs.Pronex™ 6.23 115.89 21.53 0.61
Pettibon vs.Cervitrak 5.78 99.69 19.97 0.61
Pettibon vs.Necktrac 5.65 95.39 19.53 0.61
Pettibon vs.C-Tract 5.55 91.92 19.17 0.61
Pronex™ vs.C-Tract -0.68 1.39 2.36 0.61
Significant at 95%.
the device easily and effectively. With a careful reading of
the instructions, none of the devices took more than 10
minutes to set up. In addition, Lossing Orthopedic and
Granberg International have several models to choose from
and are upgrading the devices which we used in this study.
This should improve their ease of use. Before prescribing
any device it would be wise for the physician to have the
device fully demonstrated by the dealer and tried by the
physician. This will acquaint the doctor with some of the
nuances that each of the devices had. This will enable him
or her to answer their patients questions in a more direct
fashion. Each of the devices tested had its unique charac-
teristics and quirks.
fig. 8 Average time to set up device.(min.)
The Pettibon device was a halter harness which was worn
around the skull with a strap that went under the chin. A
water bag was hung from the top of the cranium on a s-
hook. This had the tendency to bring the neck into exten-
sion and to put a lot of pressure from the nylon webbing
fabric into the jaw and forehead. It was difficult to set the
device up by oneself and once the traction session was com-
pleted, the weight could not be removed gradually to allow
the joints to accommodate to the weight of the head.
The Cervitrak by Staodyn used a preset spring tension
of 18 pounds which when uncoiled was attached to a
bracket on the halter harness which was attached to the
cranium. A foam padding headband was worn first and the
metal bracket/harness assembly was placed over it. The
bands were often tight around the skull and once the trac-
tion was applied it was an all-or-nothing phenomenon. The
device could be set up by the patient, but there was no
gradual release of the pressure when the session was com-
pleted. In this aspect it was similar to the Pettibon.
TABLE 13. One Factor ANOVA-Repeated Measures for
EMG Readings at Posterior Cervical Muscles
Mean Scheffe Fisher
Comparison difference F test Dunnett t PLSD
Pettibon vs.Pronex™ 5.81 109.76 20.95 0.6
Pettibon vs.Cervitrak 5.72 106.48 20.64 0.6
Pettibon vs.Necktrac 5.68 105.09 20.5 0.6
Pettibon vs.C-Tract 5.78 108.54 20.84 0.6
Significant at 95%.
The Necktrac by Lossing Orthopedic used a spring gauge
device which was attached to a rope-and-pulley assembly.
Once the harness was attached to the headband assembly,
the patient pulled on the rope, which allowed the traction
gauge device to be positioned in the view of the patient. A
small scale and needle on the gauge told the patient how
much poundage of traction would be applied. Once the de-
sired amount was reached, the patient pulled down on the
rope which locked it in place. We found that it took quite
a bit of effort to pull on the traction gauge. Even some of
our youngest and strongest subjects could not get past 30
pounds. In addition, it was sometimes difficult to get the
rope to lock into place. In order to accommodate the trac-
tion weight and make sure the rope locked in place at the
desired poundage, assistance was necessary from two ad-
ditional people. Also attached to the brace assembly were
two shoulder braces which curved around the trapezius
muscle. This provided a foundation for the traction force.
Some subjects indicated that as the force increased, the
metal braces began to push into the shoulders and become
painful. The tension on the device could be released grad-
ually by the patient by maintaining a slowly decreasing pull
on the rope. Lossing has since improved on this design with
a digital model which should make this problem easier to
overcome.
The C-Tract by Granberg used a winch-and-pulley sys-
tem. The patient could lie on the table attached to the har-
ness and headband assembly and then, using a rope at-
tached to a lever arm on the winch assembly, begin
repeatedly "clicking" the arm thus increasing the draw on
the traction assembly. This was repeated until the preset
tension on the dial indicator was reached. In the beginning,
the subjects had some difficulty engaging the traction be-
cause the number of "clicks" necessary to pull the rope
was very time-consuming. Once the slack was taken out of
the rope, the traction was easy to apply and could also be
released gradually.
The Pronex™ was the only device which used a totally
different concept to traction. It used two rubber cradles at-
tached by a rubber air bellows. The subject placed his or
her neck in the cradle so that the lowest cradle was on the
shoulders and the upper cradle was positioned under the
occiput. The cradle was held in place by a wide rubber band
that went across the head and was Velcroed to the other
side of the neck cradle assembly. Once the subject was
positioned in the cradle, a pneumatic-bulb was pumped by
hand to increase air pressure in the bellows. As the pressure
increased, the bellows would expand and separate the two
cradles, thus pushing the occiput away from the shoulders.
In using the device, we found that it took about 30 to 50
squeezes on the bulb to get the desired traction effect. In
addition, the ears would sometimes get pinched by the sides
of the cradle. RS Medical® indicated that the Pronex™ comes
in various sizes and that the appropriate size would elimi-
nate this ear-pinching problem. The only difficulty we had
with the device was knowing how much pressure was being
applied. Each subject was told to increase the pressure until
they reached the maximum they could tolerate. None of the
subjects had any problem reaching this level. By turning
the air-release valve, the pressure could be easily reduced,
allowing for a very controlled gradual release of traction.
All subjects tested verified that this was the easiest and
most comfortable device to use.
APPENDIX: SURVEY QUESTIONS AND RESULTS
1. The instructions were clear and easy to under-
stand.
The Pronex™ device scored the best with a score of 8.6
on a 10 point scale. The Pettibon scored the least at 5.9.
2. The device was easy to set up.
The Pronex™ scored the best with a score of 8.8. The
Granberg scored the least with a score of 5.9.
3. I did not need assistance in using this device.
The Pronex&rade; scored the best with a score of 8.8. The
Granberg scored the least with a score of 6.2.
4. It took a relatively short time to set up the de-
vice and begin using it.
The Pronex™ scored the best with a score of 8.8. The
Granberg scored the least with a score of 6.1.
5. I felt comfortable and at ease when I used the
device.
The Pronex™ scored the best with a score of 8.6. The
Pettibon device scored the least with a score of 5.7.
6. The setup was a very time-consuming step in
the process.
The Pronex™ scored the best (in this case lowest values
were better) with a score of 5.2. The Granberg was the
least with a score of 8.0.
7. The warnings and contraindications were clear
and easy to comprehend. I knew precisely whether
device was safe for my use.
The Pronex™ scored the best with a score of 7.7. The
Pettibon scored the least with a score of 5.5.
8. It was easy to measure and apply the right
amount of pressure.
The Pronex™ scored the best with a score of 7.4. The
Pettibon was the least at 6.0.
9. When the session was done, I felt that I could
release the pressure slowly to allow my joints and
muscles to re-adjust to the weight of my head.
The Pronex™ scored the best with a score of 8.4. The
Pettibon scored the least with a score of 5.1.
10. I felt that if a problem occurred during the use
of the device that I could safely and quickly get out
of the device or remove the traction pressure.
The Pronex™ scored the best with a score of 8.4. The
Pettibon scored the least with a score of 5.7.
11. I felt that if I had the need for this device, I
would feel better after using it.
The Pronex™ scored the best with a score of 7.9. The
Pettibon scored the least with a score of 6.1.
Received, August 28, 1994
Revised, March 3, 1995
Accepted, April 21, 1995
Reprint requests: Patrick P. Venditti, 2408 Himalayan Pass
Ct., Ellisville, MO 63011
References
1. Caldwell JW, Krusen EM. Effectiveness of cervical traction
in treatment of neck problems: Evaluation of various methods.
Arch Phys Med Rehabil I962;43:2I4.
2. Colachis SC, Strohm BR. Relationship of time to varied trac-
tion force with constant angle of pull. Arch Phys Med Rehabil
1966;47:353.
3. Crue BJ. Importance of flexion in cervical traction for radi-
culitis. USAF Med J 1957:374.
4. Cyriax JH. Treatment by manipulation, massage and injec-
tion. In: Textbook of Orthopedic Medicine, l Oth ed., London:
Ballier Tindall, 1982:92.
5. Geiringer SR, Kincaid CB, Rechtein JJ. Traction, manipula-
tion and massage. In: DeLisa JA, ed. Rehabilitation Medicine:
Principals and Practice. Philadelphia; J.B. Lippincott, I988:
276.
6. Jackson R. Syndrome of cervical root compression. In:
McCharty DJ, ed: Arthritis and Allied Conditions, 9th ed.
Philadelphia; Lea and Febiger, 1979:I023.
7. Saunders HD. Use of spinal traction in the treatment of neck
and back conditions. Clin Orthop 1983;t79:31.
8. Tristle HG. Conditions affecting the cervical spine. In: Good-
gold, J, ed. Rehabilitation Medicine. St. Louis; C.V. Mosby,
l 988;552.
9. Betge C, So VCk. The effect of intermittent traction on the
cervical spine. Swiss Ann 1981; VII:BS-100.
l0. Cyriax J. Trial by traction. Br Med J 1976;1:522-523.
l1. Lossing W, Boeckman P. Combining supine cervical distrac-
tion with myofascial release and soft-tissue stretch for an ef-
fective patient treatment. Am Chiro July I988:44.
12. Isaac S, Michael WB, eds. Handbook in Research and Eval-
uation, 2nd ed. San Diego, CA: EdITS Publishers, I985:I42.
13. Colachis SC, Strohm BR. A study of tractive forces and angle
of pull on vertebral interspaces in the cervical spine. Arch
Phys Med Rehabil l965;46:820-829.
I4. Jette DU, et al. Effect of intermittent, supine cervical traction
on the myoelectric activity of the upper trapezius muscle in
subjects with neck pain. Phys Ther 1985;65(8):I I73-l 176.
*Associate Professor and Past Director of Ergonomics, Chief Editor of
Occutrax, and principal of ICG, Inc.
*Director of Research.
*Professor and Director of Radiology.
*Assistant Professor and Director of Research.
JNMS: Journal of the Neuromusculoskeletal System,
Vol. 3, No. 2,Summer 1975
|
