Wood TG, Colloca CJ, Matthews R
Department of Chiropractic,
Technikon Natal, South Africa
OBJECTIVE: To determine the relative effect of instrument-delivered thrust cervical manipulations in comparison with traditional manual-delivered thrust cervical manipulations in the treatment of cervical spine dysfunction.
DESIGN: Prospective, randomized, comparative clinical trial.
SETTING: Outpatient chiropractic clinic, Technikon Natal, South Africa.
PATIENTS: Thirty patients diagnosed with neck pain and restricted cervical spine range of motion without complicating pathosis for at least 1 month were included in the study.
INTERVENTIONS: The patients were randomized into 2 groups. Those in one group received mechanical force, manually assisted (MFMA) manipulation to the cervical spine, delivered by means of a hand-held instrument (Activator II Adjusting Instrument). Those in the other group received specific contact high-velocity, low-amplitude (HVLA) manipulation consisting of standard Diversified rotary/lateral break techniques to the cervical spine. Each group received only the specific therapeutic intervention, no other treatment modalities or interventions (including medication) being used, until asymptomatic status was achieved or a maximum of 8 treatments had been received.
MAIN OUTCOME MEASURES: Both treatment groups were assessed through use of subjective (Numerical Pain Rating Scale 101, McGill Short-Form Pain Questionnaire, and Neck Disability Index) and objective (goniometer cervical range of motion) measurement parameters at specific intervals during the treatment period and at 1-month follow-up. The data were assessed through use of 2-tailed nonparametric paired and unpaired analysis, descriptive statistics, and power analysis of the data.
RESULTS: The results indicate that both treatment methods had a positive effect on the subjective and objective clinical outcome measures, no significant difference being observed between the 2 groups (P < .025). The subjective data from all 3 questionnaires showed statistically significant changes from initial to final consultations as well as from initial consultation to 1-month follow-up (P < .025). The objective range of motion measures showed statistically significant changes in the MFMA group for left and right rotation and left and right lateral flexion from initial consultation to final consultations and for right rotation and right lateral flexion from initial consultation to 1-month follow-up. The HVLA group showed only the change in left rotation from initial to final consultations and from initial consultation to 1-month follow-up to be statistically significant.
CONCLUSION: The results of this clinical trial indicate that both instrumental (MFMA) manipulation and manual (HVLA) manipulation have beneficial effects associated with reducing pain and disability and improving cervical range of motion in this patient population. A randomized, controlled clinical trial in a similar patient base with a larger sample size is necessary to verify the clinical relevance of these findings.
From the Full-Text Article:
The results of this study suggest that both instrumental and manual thrust manipulations demonstrate a positive effect in the treatment of cervical spine joint dysfunction in this patient population. Improvement appeared to carry over to the 1-month follow-up, indicating an apparent lasting benefit in these patients.
Because the results of our pilot study are not controlled, they must be interpreted with caution. They cannot be taken as proof of the clinical efficacy of manipulation for cervical dysfunction; however, the positive trends observed are sufficient to justify a call for a well-designed, randomized, controlled clinical trial in a similar patient population. The correlation between an increase in cervical rotation and a decrease in pain in both groups provides documentation of the anecdotal claims of efficacy among clinicians using these forms of SM. Because group B did not exhibit a greater effect than group A in terms of subjective (pain and disability) and objective (range of motion) findings, as had been hypothesized, we found that the 2 treatment protocols had equal effects.
There are several limitations to the study that should be discussed. The questionnaires might not have been sensitive enough to detect subtle changes in pain and disability. Likewise, the goniometer might not have been sensitive enough to detect subtle changes, inasmuch as it is calibrated at increments of 2°. The subjective intragroup data had satisfactory statistical power, as did a small proportion of the objective data. Both the remainder of the intragroup statistical analysis for the ranges of motion and the intergroup (group A versus group B) analyses for subjective and objective data exhibited unsatisfactory power. This indicates that the likelihood of committing a type II error is strong. The strength of these 2 factors (the possibilities of test insensitivity and type II error) could be reduced by increasing the sample size. This would also make trends within the samples more apparent. In future studies, efforts should be made to ensure that the groups are more homogenous with respect to level of dysfunction, duration of symptoms, age, and sex. In addition, the data collection should be done by a blinded examiner to decrease the chance of examiner bias. Finally, including a control group and a sham group would allow a greater understanding of the true clinical benefits of these manipulative procedures.
Literature review of relevant studies
Hurwitz et al  performed a systematic review of the literature to assess the evidence for the efficacy of cervical SM for the treatment of neck pain. As of 1996, they had identified 5 randomized controlled trials (Table 10), 1 cohort study, 4 case series, and 24 case reports that attempted to assess the effectiveness of cervical SM in patients with neck pain.. Table 10. Summary of randomized controlled trials of cervical spinal manipulation in treatment of neck pain (adapted from Hurwitz et al ) . Since then, other studies relevant to this work have appeared.
Randomized controlled trials
In 1996, Nilsson et al  sought to determine whether a 3-week series of SM had any lasting effect on passive CROM. Thirty-nine headache sufferers with reduced passive CROM were randomized into 2 groups to receive either HVLA cervical manipulation twice a week for 3 weeks or low-level laser therapy in the upper cervical region and deep friction massage in the lower cervical/upper thoracic region at the same frequency. The authors reported that though passive CROM increased in both groups during the trial period, there were no statistically significant differences between the 2 groups 1 week after the last treatment.
In a 1992 study, Cassidy et al  assessed the immediate effect of HVLA spinal manipulative therapy (SMT) with mobilization (in the form of a muscle energy technique) in 100 consecutive patients with neck pain. Before and immediately after the treatments, cervical spine range of motion was recorded and pain intensity was rated. The authors reported that both treatments were found to increase CROM and manipulation was found to have a significantly greater effect on pain intensity. However, after reanalyzing their data using proper analysis of covariance, which adjusted the posttreatment scores to compensate for the differences among the pretreatment scores of their patients, no significant difference in pain relief was observed between the 2 treatment groups (P = .16).  In addition, the analyses of CROM were found to be susceptible to type II error. 
Also in 1992, Nansel et al  compared upper cervical and lower cervical spinal adjustments (HVLA SM) in 69 asymptomatic subjects who demonstrated restricted passive end-range lateral flexion asymmetries of 10° or more. Goniometric cervical spine range of motion values before and 30 minutes after treatment were compared. Upper cervical adjustments were found to improve axial rotation more than treatments to the lower cervical spine, whereas lower cervical spinal adjustments were found to result in greater improvement in amelioration of lateral flexion asymmetries. The clinical relevance of these findings in patients with neck pain could not be established because of the asymptomatic status of the subjects.
In a pilot randomized, controlled clinical trial conducted in 1990, Vernon et al  reported the results of pain thresholds of the cervical spine in 9 subjects assigned to receive HVLA cervical SM or oscillatory mobilization. Patients receiving SM showed statistically higher increases in pressure pain thresholds after treatment than those in the control group receiving mobilization. Although their findings are limited by a small sample size (n = 9), the study of Vernon et al  was one of the first to compare SM in a controlled trial assessing pain threshold.
Two earlier studies had investigated the effects of cervical SM in conjunction with other medical procedures. In 1983, Howe et al  investigated the effects of cervical SM and joint injection in 52 subjects and found favorable results. In 1982, Sloop et al  found that cervical SM was associated with no significant improvements in patients’ subjective reporting of neck pain on a visual analog scale immediately after the treatment. Patients were given an amnesic dose of diazepam to provide the double-blind criteria used in the study; however, the method used does not allow generalization of the results to common clinical practice.
Other clinical trials
In 1998, Jordan et al  conducted a randomized, prospective clinical trial of 119 patients with chronic neck pain; participants received neck and shoulder exercises, physiotherapy, or HVLA SMT. Patients from all 3 groups demonstrated significant improvements regarding self-reported pain and disability on completion of the study and at 4- and 12-months’ follow-up, no significant differences being noted among the groups.
A 1997 study by Rogers  compared 6 sessions of cervical and thoracic HVLA SMT over 3 to 4 weeks with twice daily cervicothoracic stretches over the same period in 20 patients with chronic neck pain. Patients receiving SMT reported a decrease in pain levels of 44% on a visual analog scale, whereas a reduction of only 9% was reported in those performing the cervicothoracic stretches. Although only a small sample was used, the Rogers  study would seem to be supportive of the trend seen in the small but growing body of evidence—that manipulation of the cervical spine is helpful in reducing neck pain.
Comparison of spinal manipulative techniques
Little research exists with regard to comparisons of different forms of SMT or chiropractic techniques; however, one study in particular was similar to the present investigation. Yurkiw and Mior  compared HVLA SM with MFMA SM using the AAI in patients with neck pain. Each of 14 subjects was randomly assigned to one of 2 groups, evaluated by a blinded examiner, and then given one of the 2 forms of SMT. The outcome measures used were lateral flexion and a subjective pain rating. Both treatment types seemed to yield clinical improvement in lateral flexion measures and visual analog scale scores; however, the findings were not statistically significant. The authors acknowledged that the small sample size and statistically insignificant results are consistent with type II error. Noteworthy in this study, however, is the fact that the “2 ring” setting was used for the AAI treatments, which limited the excursion or depth of penetration of the Activator thrust. Yurkiw and Mior  chose this setting because they believed it to be the standard setting for adjustment of the cervical spine according to the Activator Methods protocol, but the work that they cited was dated; a maximum setting is recommended in the cervical spine except when C1 is being contacted.  Use of the maximum setting of the AAI might have yielded different clinical results in this group of patients.
In a descriptive case series  of 10 consecutive patients with neck pain being treated for whiplash, it was reported that mean pain scores and mean active CROM improved after a 6-week treatment regimen that included MFMA SMT. When followed over the course of a year, most of the patients reported stability in their improvement.
Numerous SM techniques are used in the conservative treatment of patients with cervical spine complaints. [22, 43] Concerns about deleterious effects of SM with respect to neurovascular accidents, however rare they might be, have caused the chiropractic profession to investigate alternative methods that might provide similar beneficial results in patients while involving the lowest risk possible. The risk/benefit ratio is an important consideration in any health care procedure, especially when the adverse sequelae include paralysis or death. [44-46] This is complicated by the fact that cerebrovascular screening tests have not been found to be helpful in identifying those at risk. 
Conflicting reports have appeared with regard to cervical rotational SM procedures’ being associated with an increased incidence of cerebrovascular incidents (CVIs) and cerebrovascular accidents (CVAs). [48-50] After a detailed search of the literature through 1993, it was reported by Haldeman et al  that “the literature does not assist in the identification of the offending mechanical trauma, neck movement, or type of manipulation precipitating vertebrobasilar artery dissection or the identification of the patient at risk”. However, studies beginning in 1996 have found cervical spine rotational maneuvers to be associated with CVAs and CVIs after SM more strongly than other “nonrotational” techniques. [4, 48, 49, 51] The purported safety of using MFMA-driven SM with a device such as the AAI is thought to be due to the prone neutral positioning of the patient during the SM procedure (nonrotation) combined with the controlled, repeatable force of the thrust in the joint plane line. 
Currently, investigations have begun to differentiate mechanical and physiologic responses of SMT. [18, 43, 53-58] Neurophysiologic models theorize that SMT stimulates the somatosensory system and can subsequently evoke neuromuscular reflexes. [53, 59-61] Although there is little research pertaining to the cervical spine in this regard, MFMA SMT has been found to elicit significant neuromuscular reflexes in the erector spinae musculature in patients with LBP. [57, 62] Traditional SMT has also been found to elicit neuromuscular responses in the cervical spine in asymptomatic subjects.  Such mechanical and neurophysiologic studies indicate that joint manipulation might have both direct and indirect clinical benefits. Beneficial effects of SMT have been thought to be associated with mechanosensitive afferent stimulation and presynaptic inhibition of nociceptive afferent transmission in the modulation of pain, [63, 64] inhibition of hypertonic muscles, [18, 53, 65] and improved functional ability. [1, 66, 67] Such theories must be substantiated by well-designed investigations to verify the clinical relevance of these mechanisms
This study demonstrated that HVLA manual thrust manipulations to the cervical spine show no benefit over MFMA instrumental thrusts with respect to objective and subjective clinical findings in patients with chronic cervical spine dysfunction. Furthermore, there is sufficient clinical and statistical evidence (except for some ranges of motion) to suggest that both treatment protocols had an effect on the participants in this study, but these results would have supported the use of MFMA and HVLA manipulation for cervical spine dysfunction only if a control group had been included in the investigation. Visual inspection of the data suggests that both treatment groups experienced a decrease in pain and disability and an increase in CROM after treatment and at the 1-month follow-up. Such findings are encouraging and form the basis for a larger-scale, randomized, controlled clinical trial to further investigate their clinical relevance. As technique options become available to clinicians, research must continue to identify SM techniques that maximize therapeutic outcomes while minimizing patient risk