J Manipulative Physiol Ther. 2016 (Jun); 35 (5): 319–329 ~ FULL TEXT
Lindsay M Gorrell, MChiro, MRes, PhD, Kenneth Beath,
Roger M Engel, DO, DC, PhD
Department of Chiropractic,
Sydney, NSW, Australia.
OBJECTIVE: The purpose of this study was to compare the effects of 2 different cervical manipulation techniques for mechanical neck pain (MNP).
METHODS: Participants with mechanical neck pain (MNP) of at least 1 month's duration (n = 65) were randomly allocated to 3 groups:
(1) stretching (control)
(2) stretching plus manually applied manipulation (MAM), and
(3) stretching plus instrument-applied manipulation (IAM).
MAM consisted of a single high-velocity, low-amplitude cervical chiropractic manipulation, whereas IAM involved the application of a single cervical manipulation using an (Activator IV) adjusting instrument. Preintervention and postintervention measurements were taken of all outcomes measures. Pain was the primary outcome and was measured using visual analogue scale and pressure pain thresholds. Secondary outcomes included cervical range of motion, hand grip-strength, and wrist blood pressure. Follow-up subjective pain scores were obtained via telephone text message 7 days postintervention.
RESULTS: Subjective pain scores decreased at 7–day follow-up in the manually applied manipulation (MAM) group compared with control (P = .015). Cervical rotation bilaterally (ipsilateral: P = .002; contralateral: P = .015) and lateral flexion on the contralateral side to manipulation (P = .001) increased following MAM. Hand grip-strength on the contralateral side to manipulation (P = .013) increased following IAM. No moderate or severe adverse events were reported. Mild adverse events were reported on 6 occasions (control, 4; MAM, 1; IAM, 1).
CONCLUSION: This study demonstrates that a single cervical manipulation is capable of producing immediate and short-term benefits for mechanical neck pain (MNP). The study also demonstrates that not all manipulative techniques have the same effect and that the differences may be mediated by neurological or biomechanical factors inherent to each technique.
KEYWORDS: Cervical Vertebrae; Chiropractic; Manipulation; Neck Pain; Randomized Controlled Trial; Spinal
From the Full-Text Article:
The annual prevalence of neck pain is estimated to range from 30% to 50%, with reports of lifetime and point prevalence values approaching those of low back pain. [1–6] Mechanical neck pain (MNP) is defined as nonspecific pain of nonpathological origin occurring in the cervical spine. [7, 8] A common approach to managing MNP includes cervical spine manipulation. [8, 9] Although high-velocity, low-amplitude (HVLA) cervical manipulation has been shown to be effective for treating MNP [9, 10] and is included in several clinical practice guidelines, [11–13] the optimal manipulative technique for treating this condition remains ambiguous.  HVLA manipulation can be delivered manually (manually applied manipulation [MAM]) or by instrument (instrument-applied manipulation [IAM]). However, there is no clear evidence to support one approach over the other. [9, 15–23] MAM is commonly used and involves the manual application of a force aimed at moving a joint beyond its physiological range of motion (ROM) without exceeding the anatomical limit. [16, 24] By contrast, the delivery of a manipulative force in an IAM does not rely on moving a joint beyond its physiological ROM to achieve an effect. 
This difference in approach has not been adequately reflected in reports of change following spinal manipulation. [26–30] Three studies comparing the effectiveness of MAM and IAM for the treatment of MNP reported both approaches to be equally effective. [31–33] However, the quality of these studies was poor with inadequate sample sizes, lack of a control group, and heterogeneous methodologies, detracting from the validity and generalizability of the results.  This is in contrast to the findings of the largest study to date comparing MAM, IAM, and usual care for the treatment of low back pain which reported that, in a population of 107 participants, MAM provided greater short-term reductions in self-reported disability compared with IAM and usual care. 
Remote effects following HVLA spinal manipulation, that is, effects which occur in tissues not directly related to the area where the intervention was applied, have also been reported in the literature. [36, 37] The hypothesis that there is a connection between the response of the autonomic nervous system and pain perception following spinal manipulation has been investigated by a number of researchers. [36, 38, 39] Reports of changes in skin conductance, respiratory rate, blood pressure, and heart rate in healthy populations following mobilization or manipulation of specific areas of the spine support this hypothesis. [39, 40]
In addition to responses generated by the autonomic nervous system, cervical spine manipulation has also been associated with changes in the somatic nervous system. [41, 42] Studies investigating the effect of cervical manipulation on lateral epicondylalgia have described an increase in hand grip-strength, [43–45] whereas other studies have reported excitatory effects on motor activity. [46–48]
The aim of this study was to determine whether a single application of HVLA cervical manipulation (MAM or IAM) affected MNP and, if so, whether the effect was the same for both types of manipulation.
The findings reported in this study show that a single application of cervical manipulation increases cervical ROM and decreases subjective pain levels in people with MNP. The study also shows that cervical manipulation produces remote effects. However, these effects are noticeably different for each type of manipulation. MAM produced immediate increases in rotation bilaterally and lateral flexion on the contralateral side to manipulation compared with IAM and a decrease in subjective pain levels 7 days postintervention when compared with control. IAM did not produce equivalent changes. Our results support the findings of a recent large study comparing MAM, IAM, and usual care in a cohort with low back pain which reported greater reductions in short-term self-reported disability and pain scores following MAM.  However, our results contradict the findings from other smaller studies which reported that MAM and IAM were equally effective in the treatment of MNP. [31–33] In the absence of a control group, it is possible that performance bias may have affected the results from these smaller studies. [36, 63–66] In addition, given that these studies reported on small sample sizes (14–47 participants), it is possible that a series of type II errors may have occurred, resulting in incorrect acceptance of the hypothesis that there is no difference between MAM and IAM. [67, 68] Given the quality of previous studies investigating manipulation for MNP, we will compare our results with studies of higher quality that investigated the effects of spinal mobilization because the forces used in those studies are similar to those used in our study. [69, 70]
Subjective Pain Levels (VAS and NPRS)
Our results are consistent with previous reports that MAM is more effective than mobilization in reducing subjective pain levels for MNP. [10, 30, 71, 72] As in our study, time appears to be a factor, with these studies reporting improvements at 2,72 4,10 and 771 days postintervention. In contrast, Martinez-Segura et al  reported an immediate decrease in subjective pain levels following a single cervical HVLA manipulation compared with mobilization in participants with MNP.
Our results contradict the findings of 2 studies comparing the effectiveness of cervical MAM with mobilization in patients with neck pain. [15, 73] However, this contradiction may be explained by the heterogeneous nature of their designs compared with our study. In the first trial, patients with radiculopathy were included as well as treatment to the thoracic spine,  whereas in the second trial, treatment to the thoracic and lumbar regions was included, randomization only occurred after several treatments, and the type of MAM and mobilization procedures was not reported. 
As none of the groups in the current study reported changes that were above the MCID for either VAS or NPRS, the results should not strictly be interpreted as clinically significant. As our cohort displayed relatively low pain levels at baseline, it is possible that a “floor effect” may have contributed to this lack of clinical significance. [74, 75] This floor effect may have been avoided by excluding participants who reported a baseline pain level less than 3 of 10.  Furthermore, interpretation of subjective pain scores can vary between individuals. For example, some patients interpret the minimum end point on the scale (labeled “0”) as indicative of “normal” or “manageable” pain. These patients often exclude the lower half of the rating scale, considering such levels of pain as outside their experience. Interpretation of subjective pain scales in this way can result in distorted scores because these patients only use part of the range, that is, 6–10, leading some authors to question the validity of subjective pain scales. [76–80]
In addition, based on the change scores reported immediately postintervention, it is hypothesized that the MCID may have been achieved at 7–day follow-up had a second manipulation been applied such as would usually occur in clinical practice.
Pressure Pain Threshold
Our finding of no between-group differences in PPT immediately postintervention is consistent with results from other studies that found no differences in PPT following high -and low-force mobilization in people with chronic, nonspecific neck pain  or whiplash.  Furthermore, none of the reported changes in the current trial were above the MCID for PPT. This finding is also consistent with Martinez-Segura et al  who reported nonsignificant increases in PPT following both cervical and thoracic manipulation for MNP. As PPT is a self-reported measure of pain, it is possible that our results may have been influenced by participants who had varying pain thresholds, specifically where a sensitive participant may have reported pain earlier than a more stoic participant. In addition, the control group was not blinded, and it is possible that participants in this group may have had a falsely lowered pain threshold.
The immediate increase in cervical rotation and lateral flexion following MAM reported in this study contradicts findings from other studies where no change to cervical ROM was reported following cervical mobilization or manipulation in populations with MNP. [10, 30] Furthermore, 2 recent systematic reviews concluded that there was uncertainty as to whether spinal manipulation improved cervical ROM. [82, 83]
In explaining our findings, consideration must be given to the fact that we used an MAM technique which facilitated distraction and displacement of the cervical spine. As preload manipulative forces are capable of eliciting changes in paraspinal muscle activation, [23, 84–86] it is suggested that the chosen technique exerted a preload force that achieved this activation, thereby facilitating a change in cervical ROM. This may explain the lack of change to cervical ROM seen subsequent to IAM, as different preload forces are generated using this technique.
This is the first study to report immediate increases in hand grip-strength following IAM and contradicts other reports of changes following manipulation. [41, 42, 44, 45] Notwithstanding this, the finding of increased hand grip-strength on the contralateral side to manipulation is supported by Humphries et al  who reported increases in a population of asymptomatic basketball players. Furthermore, other authors have reported an immediate increase in hand grip-strength on both the affected side (ipsilateral) [44, 45] as well as bilaterally  following cervical MAM. The current findings are congruent with previous studies reporting excitatory effects on motor activity subsequent to spinal manipulation. [46–48, 87] More specifically, Dunning and Rushton  demonstrated that a single cervical manipulation can elicit an immediate increase in resting electromyographic activity in an area distal and not directly connected to the cervical spine. Other studies investigating the effects of cervical manipulation on lateral epicondylalgia have described both sensory and motor changes including an increased hand grip-strength following manipulation. [44, 45, 89] It has also been reported that multiple sessions of cervical manipulation may produce a cumulative effect on hand grip-strength, supporting the concept of a dose-response mechanism. [41, 44] As the increases in hand grip-strength reported in this trial were not above the MCID, any interpretation of their clinical significance should be restrained.
Wrist Blood Pressure
Previous studies have reported mixed results when measuring changes in blood pressure following spinal manipulation. [36, 90–92] Two systematic reviews reflect this observation, with one reporting that spinal manipulation produced a nonsignificant, minimal clinical effect on blood pressure,  whereas the other reported strong evidence for a positive change in blood pressure in healthy populations following manipulation.  To the authors’ best knowledge, this is the first study to investigate the effect of cervical manipulation on blood pressure in a population with MNP.
Although the absence of any moderate or severe adverse events following 21 MAMs and 22 IAMs in a young population with MNP is welcomed, it should not be interpreted as an unequivocal endorsement of the safety of cervical manipulation. Reports of this kind are evidence of the “relative” safety of cervical manipulation and add to the body of evidence in the field. Furthermore, reporting the rate of minor adverse events (4.7% in the current study) promotes evidence-based discussion on the risks associated with this type of intervention.
There are a number of limitations which should be considered when interpreting the results of this study. Firstly, the trial was designed as a “proof of concept” with the choice to use a single manipulation as the intervention driven by the desire to directly measure dose-response, an approach previously recommended by several authors. [84–86, 94] The results should therefore not be interpreted as indicative of clinical practice.
Secondly, as participants in the trial were young with low baseline levels of pain, the results may not be generalizable to older patients or those with other cervical complaints such as neck trauma or radiculopathy.
Thirdly, we attempted to reduce interoperator inconsistencies by using a single clinician for each type of manipulation; therefore, it is possible that similar studies using multiple practitioners may have difficulty replicating our results.
Fourthly, the reliability and validity of VAS [53, 75] and NPRS  have been reported separately; however, this does not extend to inclusion in the same analysis. To maintain consistency across measures, the average of a multiple-question NPRS could have been used at all time points to report subjective pain levels. In addition, it is possible that as numerous multiple comparisons were performed, a type I error may have occurred.  Furthermore, as there was no interexaminer validation of the diagnosis, it is also possible that the initial diagnosis may have been incorrect and therefore influenced the results. Notwithstanding, a recent study supports the use of a targeted physical examination involving pain provocation with manual palpation for accurate diagnosis of MNP. 
This study demonstrates that a single cervical manipulation is capable of producing both immediate and short-term benefits for MNP. The study also demonstrates that not all manipulative techniques have the same effect. The results reported in this study are consistent with the hypothesis that the biomechanical characteristics of different spinal manipulation techniques may be responsible for varying clinical effects. However, the results are not definitive, and further research investigating the nature of these changes is warranted.
A single application of MAM increased rotation and lateral flexion movements
and decreased pain levels in people with mechanical neck pain.
A single application of IAM increased hand grip-strength on the contralateral
side to manipulation.
There was no change in PPT following either type of manipulation.
Following 21 MAMs and 22 IAMs, there were no moderate or severe adverse events
and 2 mild adverse events reported in this trial.