Most traction devices that operate on the lumbar spine use a belt or other harness that supports the user’s pelvis. This belt is attached to a tension device or weight by way of a cable. The combination pulls on the lower trunk of the user, applying tension in an attempt to lengthen the spine. Other variations
include using the person’s body weight as the weight. These are typically inversion devices where the user is positioned partially or completely inverted and supported by their legs or hips. The mass of the upper body provides tension to decompress or lengthen the spine.

A more subtle type uses spinal manipulation to decompress a portion of the spine. Fig. 1 shows a torso in a prone (face down) position with the vertebra in a substantially neutral position. When the ilium is elevated, typically by applying a vertical force on the anterior iliac crest, the pelvis will rotate posteriorly. This is illustrated in Fig. 2. The pelvic rotation provides greater disk spacing in the posterior aspect of the lumbar vertebrae. The posterior aspect of the vertebrae is adjacent to the nerve cord. By increasing the posterior spacing of the vertebrae, the disks between the vertebrae are placed in mild traction. A bulge in a disk in this area would tend to be flattened and pull away from the nerve cord. This would reduce pressure in the area of the nerve cord and potentially enable pain relief.

Methods:
This study examines the resultant pain relief by positioning a patient with low back pain in a prone, pelvic elevated position. Subjects were identified with low back pain. A designation of hyperlordosis (excessive lumbar curve) or hypolordosis (reduced lumbar curve) was not made in an attempt to generalize the population of individuals with low back pain. Subjects with low back pain participated on a voluntary basis. A total of 14 subjects (7 male and 7 female) participated in the study. The study was performed at a chiropractic office in the United States. The subjects that participated in the study did so after their chiropractic session. The subjects were not screened other than they experience lower back pain to some degree.

(Fig. 3 Control chair (left) and ErgoLounger (right) shown prior to use.)

Subjects were asked to rate their low back pain using a Visual Analog Scale (VAS) under five conditions. An unmarked sample of the VAS is attached in the appendix. The subjects marked the VAS without any interference or coaching by the test administrators. The first condition was standing in a neutral position with arms at their sides (pre-test (PRE)). This was done after the chiropractic treatment that day. The second pain rating was preformed while laying face down (prone) on either a traditional flat folding lounge chair (control chair (CC)) or a declining chair with pelvic support (ErgoLounger, DFE Enterprises, Inc., New York) (EL). The control chair is a common folding lounge chair. No structural changes were made to the control chair so that the chair would simulate a traditional device commonly found in retail outlets throughout the country.

The subjects were asked to lie face down and relax for 30 seconds. After the 30-second time period they were asked to rate their pain by marking the VAS. The subjects were then asked to stand up in a neutral position similar to that of the pre-test and rate their pain after using that chair (P-CC or P-EL). Follow the same procedure on the other chair, either the CC or the EL, the subjects rated their pain level after 30 seconds on the second chair and again after using that chair. Subjects were alternatively assigned
Chair Report 4/6/2004 Page: 4 of 4 to the order of the testing procedure (EL or CC first) upon entering the room. When the EL was used first, the P-EL (post ErgoLounger) was used as the pre-CC and likewise the P-CC (post control) was the pre-EL when the CC was tested first.


The subjects came into the testing room after their chiropractic appointments throughout the day. The order of the subjects was unknown to the test administrators prior to the subjects entering the room. This was done to remove any bias or combined or carryover effect of pain relief from one chair to the other. The subjects were assigned a number to identify the subject and the order of testing was also noted. The subjects were not compensated for participation in the study and the subjects were not informed as to which device was the control chair and which was the experimental chair. The test administrators were unknown to the subjects and the subject to the test administrators prior to the study. The management at the chiropractic office where the test was conducted, arranged the subject’s participation in the study

Results:
The Visual Analog Scale (VAS) sheets were measured and normalized by the total length of the line. The resulting values were multiplied by 100 to give a relative percentage of maximal pain. Initial pain values were recorded from 0 to 72.6. The randomized order intended to minimize any pre-pain levels of either chair test. The mean pain level prior to use of the ErgoLounger (EL) was 21.97 and the mean level prior to using the control chair (CC) was 21.44. An unpaired t-test was performed and there was no significant difference between the two values (p>0.500).

The mean pain ratings of each condition are shown in Fig. 7. The bars show the mean values of the 14 trials for each condition. The error bars indicate one standard deviation above the mean. This size of this value is relevant in that it illustrates the great diversity of pain levels reported by the subjects. There were no statistically significant differences between any of the groups. This was determined by a one-way analysis of variance (ANOVA) run on the data. The great variance in pain levels prior to the study and the small number of subjects explains the lack of significant difference. A trend is suggested. The actual mean values with their standard deviations are recorded in Table 1.

Mean pain values displayed as a percent of maximum

A more telling evaluation of the differences between the EL and CC would be the change in pain level from prior to use to during use. In order to do this, the prior to use values must be clarified. In 50% of the cases the Pre-Test values were prior to use values for the EL and 50% were for the CC. The other
Chair Report 4/6/2004 Page: 6 of 6 50% for each were the post-test (P-EL and P-CC) values for the other chair. The data was evaluated in this manner to establish the actual Pre-Current-Post values for both the EL and the CC. The results are graphically illustrated in Fig. 8.

Changes in back pain prior, during and after using CC and EL

The pain levels prior to use of the CC and EL were 21.44 and 21.97 respectively. The values were not statistically different when evaluated by an unpaired t-test (p=0.959). The mean back pain experienced by the users of the CC was substantially greater than the pre-test mean value. The mean pain value was increased by 12.36 points to reach a value of 33.80. After using the EL, the subjects reported a lower pain reading of 12.11 points. This is a reduction of 9.86 points. The change in pain values, as in the tested value on the respective chair minus the pre-test values, was tabulated for each subject on each chair. These data sets were evaluated. The mean reduction in pain observed with the EL was 9.86 points and the same subjects experienced a mean increase in pain level of 12.36 points. These data sets were evaluated by an unpaired t-test to validate a statistically significant difference (p=0.002). The actual raw data (not the change) were also evaluated. The EL value of 12.11 points was statistically significant from the CC value of 33.79 points. This was confirmed by both an unpaired t-test (p=0.024) and the Mann-Whitney Rank Sum Test (p=0.006).

The change in value from Pre-test to test and pre-test to post-test were also evaluated in the user of both chairs. The reader will notice the mean post-test value of the CC was greater than mean the pre-test value and the mean post-test value of the EL was lower than the mean pre-test value. A paired t-test was used to evaluate whether the single treatment of each chair independently was significant relative to their respective pre-test values. The EL values were statistically significant (p=0.046) from pre-test values to tested (on the chair) values. The pre-test to post-test values for the EL were not statistically significant
Chair Report 4/6/2004 Page: 7 of 7 (p=0.112). In a similar manner, the pre-test values of the CC were determined to be statistically significant (p=0.019) where the pre-test to post-test values were not significant (p=0.154).

The change noted from pre-test to test between the two conditions is quite evident. This is further

Change in pain relative pain level with EL and CC illustrated in Fig. 9. The error bars are one standard deviation plus and minus the mean. As is seen, the standard deviation in both cases are greater than the respective mean values. This is evident in that over 33% of the users reported no pain prior to use (pre-test). It is interesting to note that while none of those reporting zero pain in the pre-test reported pain in the EL condition, 100% of those subjects reported pain in the CC condition.

Conclusions:
In this study subjects positioned prone on the ErgoLounger (EL) for 30 seconds showed a statistically significant reduction in low back pain as compared to pre-testing values. The EL condition noted a 44.9% reduction in the mean pain levels from a standing condition prior to testing. The post-test (standing) values after using the EL were lower than the pre-test values by 27.6%, but the findings were not statistically significant. This greatly contrasted the findings of the control chair (CC). The control chair noted a statistically significant increase in mean back pain value from the pre-test to the test values. The increase in mean pain value was 57.6% greater than the mean pre-test value. The mean post-test value of the users with the CC was greater than the mean pre-test value. The mean post-test value increased by 14.0% from the mean pre-test value, though this was not statistically significant. The difference between the mean pain level values of users in a prone position while on the two chairs (CC and EL) was also statistically significant. This offers supporting evidence that the EL is more effective in acute back pain relief than a traditional folding lounge chair (CC). The study also offers evidence that the EL may be effective in the reducing low back pain.

2. Braza, D. W. (2003) Three Key Components to Back Injury Prevention. (interview article) Medical College of Wisconsin. http://healthlink.mcw.edu/article/1012425715.html

3. Day, JW, et al. (1984) Effect of pelvic tilt on standing posture. Phys Ther. Apr;64(4):510-6.

4. Scannell, J.P. and McGill, S.M. (2003) Lumbar posture—should it, and can it, be modified? A study of passive tissue stiffness and lumbar position during activities of daily living. Phys Ther. Oct;83(10):907-17.

5. Harrilson D.E. et al. (2002) Changes in sagittal lumbar configuration with a new method of extension traction: nonrandomized clinical controlled trial. Arch Phys Med Rehabil. Nov;83(11):1585-91.

6. Janke, A.W., et al. (1997) The biomechanics of gravity-dependent traction of the lumbar spine. Spine. 22(3):253-260.

7. Tsuji, T. et al. (2001) Epidemiology of low back pain in the elderly: correlation with lumbar lordosis. J Orthop Sci. 6(4):307-11.