Does the adjustment cavitate the targeted joint? an investigation into the location of cavitation sounds☆

Introduction 

In cases of mechanical lower back pain, the chiropractor identifies lumbar or sacroiliac dysfunction using relevant clinical tests. Once the symptomatic joint is identified, it is further examined for restricted motion using motion palpation.1 This information is then used in selecting an appropriate adjustive technique, which is aimed at restoring normal motion patterns to the affected joint.2 Accuracy in delivering the adjustive force to the targeted joint therefore seems to be essential. However, the specificity of an adjustment seems to be limited by the fact that each lumbar vertebra is involved in 4 facet joints3; thus, any one or all of these may be cavitated during adjustment of the lumbar spine. The limitations of sacroiliac joint adjustment seem even greater due to strong ligaments surrounding this joint,4 suggesting that significant separation, as is associated with cavitation, is at least very difficult.

The debate concerning the difference between an adjustment and manipulation also suggests that specificity is an attribute more inherent to the chiropractic adjustment than the manipulative techniques of physiotherapy or osteopathy.1 Both procedures are capable of producing cavitation sounds,5, 6 but manipulation makes use of longer lever movements, which induce movement over a larger area than that supposedly affected by an adjustment, which uses short levers to specifically move only one joint without affecting neighboring articulations.1, 7

According to some authors, cavitation is a sign that the procedure has been performed correctly and thus will have the desired therapeutic effect.8 However, other authors contest the clinical significance of the cavitation sound.9 It is believed that the cavitation sounds are produced when the articular surfaces of the joint are sufficiently separated during an adjustment.10, 11, 12 Brodeur13 reviewed the pertinent literature of this phenomenon and reported that the mechanism of sound wave production is caused by a sudden decrease in intra-articular pressure when the joint is distracted by the adjustive thrust, resulting in the release of mainly carbon dioxide gas from the synovial fluid. The work by Sandoz5 showed that once a joint had been cavitated, it displayed an increased range of motion. This suggests that cavitation is associated with a beneficial result in terms of improved range of motion, but it is still unclear whether this is a result of the cavitation. There are certainly other methods of increasing joint range of motion, but it may be that the adjustment associated with a cavitation is one method that achieves this result almost instantaneously.

Possibly, the first recordings of cavitation sounds in the spine were conducted by 2 dentists, Woods and West.14 With the help of chiropractors, they recorded cavitation sounds of cervical, thoracic, and lumbosacral joints during manipulation and compared them with temporomandibular joint (TMJ) sounds. They concluded that there was little reproducibility of the wave form from one manipulation to the next and that the sounds produced by spinal manipulation were significantly different from TMJ sounds.

Herzog et al15 studied 20 symptomatic patients to analyze the cavitation sounds with a posterior to anterior thrust on the transverse process of the T4 vertebra, assessing the sounds using an accelerometer. They concluded that the signals were typical of a triphasic wave form and approximately 20 milliseconds in duration. Other authors, such as Gal et al,16 have, however, found that cavitation signals were more likely to be biphasic. This may be why Woods and West14 concluded that wave forms were not significantly reproducible in terms of wave form, but it also suggests that further studies are needed to investigate the wave forms associated with cavitation sounds.

Gal et al16 studied the correlation of cavitation sounds (accelerometers placed relative to the T11 and T12 vertebrae) and bone movement (embedded bone pins monitored by high-speed cinematography) in unembalmed postrigor cadavers. They found that a thrust delivered posterior to anterior to the right transverse process of the T12 vertebra produced a cavitation sound that was associated with a relatively larger degree of left lateral translation of the T12 vertebra, as compared with the T11 vertebra. They found that triphasic waves occurred during attempted thrusts without cavitation but that thrusts associated with a cavitation were associated with a biphasic wave form.

Reggars and Pollard17 investigated the relationship between the side of head rotation during rotary cervical spine manipulation and the side of joint “cracking.” Forty-seven (94%) of the 50 subjects tested exhibited cracking sounds on the same side as the rotational direction, while 2 subjects exhibited bilateral sounds and 1 subject exhibited a contralateral cavitation. This study analyzed results based on the question of the specificity of the manipulative thrust—does cavitation take place at the targeted joint?

Brodeur's13 review article discusses the mechanics of cavitation, and he proposes that the elastic recoil of the synovial capsule after the input of an adjustive thrust brings about the cavitation phenomenon. His MEDLINE database search did not seem to provide information relevant to the question posed in this study: Do adjustments cause cavitation at the specifically targeted joints?

In further work by Gal et al,18 a similar question was addressed in terms of the location of vertebral movement associated with the adjustive thrust. They used high-speed cinematography of bone pins implanted into the spinous processes of T10, T11, and T12 to show that during posterior to anterior thrusts to the right transverse processes of either T10, T11, or T12, significant movement was measured mainly at the targeted and immediately adjacent vertebrae. Although their earlier work16 did show correlation of the cavitation sound location and movement of the targeted bone, it is difficult to ascertain anything definite from this research due to the small sample size and the possibility that vertebral movement may be considerably different in postrigor cadavers than in vitro subjects.

The aim of this investigation was therefore to evaluate the effects of 2 different chiropractic adjustive techniques performed on the fifth lumbar vertebra and the sacroiliac joint in terms of the location of the cavitation sounds produced to determine whether these techniques resulted in significantly different areas of joint cavitation.

The first objective was to determine the effect of the L5 pull adjustment in terms of the location of the cavitation sound produced. The second objective was to determine the effect of the lower sacroiliac adjustment in terms of the location of the cavitation sound produced. The third objective was to compare the results of objectives 1 and 2 to determine whether these techniques resulted in significantly different areas of joint cavitation.

Methods 

Students at the Department of Chiropractic at Technikon Natal were asked to volunteer as participants for the study. All volunteers had to comply with the following criteria:

Thirty-six volunteers were found to be eligible and entered the study after their informed consent had been obtained. However, 6 subjects were excluded during the course of the study for the following reasons:

The thirty subjects used in the study were randomly divided into 2 groups of 15 subjects each. Simple randomization was achieved by sequential drawing of group allocation from a hat. Group 1 received the lumbar spinous hook adjustment,20 and group 2 received the lower sacroiliac adjustment.20

At the initial consultation, examination of the student volunteers was undertaken to establish compliance with the acceptance criteria. This was followed by an experimental consultation within a few days, at which time 8 microphones (mounted in rubber grommets) were secured to the skin over the facet joints of the fourth and fifth lumbar segments and the posterior superior iliac spine using adhesive tape. Placement was initially guided by palpation of anatomical landmarks.

Radiographic examination was then used to confirm or adjust the microphone positioning to ensure optimum placement relevant to the target joint location (Fig 1). The microphone leads were then connected to a computer with a filtering and amplification unit. A Pentium 90 MHz processor (Intel, Santa Clara, Calif) was used in conjunction with an Intelligent Instrumentation data acquisition board (PC 20450P-10). An amplifier was also designed to provide the correct levels of signals. The sound signals were sampled via an analog to digital converter at a frequency of 12.5 kHz. Software was written using the Visual Designer package, which implements fourth generation block programming. The spectrum sound of the joint cavitations was found to be 250 Hz to 750 Hz. Above this region, the signal-to-noise ratio was found to be quite high, and for this reason, the selected spectrum was used. Mathematical models for image representation of frequency signals, called Fast Fourier Transforms, were used to convert the input signals to a frequency spectrum. The device provided an 8-channel output (1 per microphone), which was graphically displayed on the screen as numerical values, bar graphs, and spectrum plots. The channel recording the highest amplitude was selected as the microphone–joint location where the cavitation sound originated, even if multiple channels recorded a signal.

View Large Image Download to PowerPoint

Fig 1. Lumbar radiograph showing microphone placement.

All subjects were positioned in the left lateral recumbent position, and the respective adjustments were executed. In some cases, several attempts were executed to achieve the audible cavitation sounds.

The sample sizes of 15 subjects indicated the use of nonparametric statistical tests. The tests included the Mann-Whitney U test for intragroup analysis and the Wilcoxon signed rank test for intergroup analysis. A 95% confidence interval was used and any subsequent P values of less than .05 were considered significant.

Results 

The Mann-Whitney U test indicated that no significant differences existed (P = .210) within each respective group. The Wilcoxon signed rank test resulted in acceptance of the null hypothesis for both groups (L5 pull: P = 0.188; lower sacroiliac: P = .355), indicating that no particular joint was cavitated frequently enough to signify specificity of the 2 adjustments.

Discussion 

The intragroup comparison indicated that the 2 different methods of adjustment did not produce cavitations at significantly different locations. This, at least, indicated that the adjustments were respectively delivered within the statistical range of uniformity.

The intergroup analysis indicated that no particular joint was cavitated frequently enough to be associated with either of the adjustments used. This contradicts authors such as Lewit,8 Schafer and Faye,2 and Sandoz5 in terms of the importance of obtaining a cavitation sound during an adjustment, as the sounds obtained are just as frequently produced by nontargeted joints in the area. However, it does lend support to Herzog et al,15 who suggested that increased reflex electromyographical activity of the surrounding muscles was due to the speed of the adjustment and not the cavitation sound.

Of interest in terms of intergroup differences was the higher average number of cavitations produced per individual during the lower sacroiliac adjustment (4.22 out of 8) than during the L5 pull adjustment (3.37 out of 8). This may be due to the contact on the ilium acting as a longer lever on the lumbar spine rather than a short lever for the sacroiliac joint and should therefore be considered a manipulation than a specific adjustment. Another aspect of interest is that the microphone that recorded the highest number of signals during the L5 pull adjustment was placed over the L3-4 right facet (the upper side during the adjustment). This may have been due to lumbar biomechanics and/or the individual technique of the researcher (Fig 2).

View Large Image Download to PowerPoint

Fig 2. Number of cavitations per microphone location.

The microphone that recorded the highest number of signals during the lower sacroiliac adjustment was placed over the right L5-S1 facet joint (upper side). This suggests that what is actually a lumbar sacral cavitation may often mistakenly be identified as a sacroiliac joint cavitation. This seems reasonable when one considers the strong ligaments which bind the sacroiliac joint.19

It was decided prior to execution that any adjustment resulting in 2 or more channels exhibiting very similar amplitudes that were higher than all other channels would indicate cavitation at both those locations. Although no such situation was found during the study, multiple channel signals were the norm, with similar amplitudes happening frequently. These signals were, however, never the highest amplitude signals.

Conclusion 

This study indicates that neither adjustment was associated with any of the targeted joints frequently enough to be of statistical significance. Moreover, the sounds recorded during the 2 adjustments did not differ significantly.

No conclusive specificity was determined with respect to the joint-targeted techniques tested.