J Manipulative Physiol Ther 2008 (Sep); 31 (7): 553—557 ~ FULL TEXT
Donald R. Murphy, DC, Jacqueline L. Beres, DC
Rhode Island Spine Center,
Pawtucket, RI 02860, USA.
OBJECTIVE: Cases have been reported in which radiculopathy or myelopathy secondary to herniated disk has occurred after cervical manipulation. In each case, it is not possible to determine whether the neurologic symptoms and signs were directly caused by the manipulation or whether they developed as part of the natural history of the disorder. The purpose of this article is to report a case in which a patient with radiculopathy secondary to herniated disk was scheduled to receive manipulation but just before receiving this treatment developed acute myelopathy.
CLINICAL FEATURES: A patient with arm pain and numbness was referred by a neurosurgeon for nonsurgical consult. He had a large C5-6 disk herniation with no signs or symptoms of myelopathy. He was determined to be a candidate for nonsurgical intervention, including manipulation. Manipulative treatment was planned for the second visit.
INTERVENTION AND OUTCOME: Ten days after the initial visit, and before any manipulative treatment being rendered, the patient developed symptoms suggestive of myelopathy, which were later determined on examination to be related to acute myelopathy secondary to the disk herniation.
CONCLUSION: Herniated disk in the cervical spine can progress to myelopathy as part of the natural history of this condition. Because of this, any interpretation of myelopathy that occurs after cervical manipulation, or any other procedure, must be made with caution.
From the Full-Text Article:
It is important for spine specialists to be aware of the diagnosis of CM. In addition, it is important to be aware of the possibility in patients with disk herniation of progression from radiculopathy to CM. The typical initial symptoms of CM are neck and arm pain, upper extremity paresthesias (often global and nondermatomal), and changes in gait and balance.  It is important in particular to question the patient with regard to balance and gait problems because these are among the few typical initial symptoms that are specific to CM. Decline in fine motor control may be seen as well as uni- or bilateral arm or leg weakness and changes in bowel/bladder function.2 On examination, one might find “myelopathy hand,” that is, wasting of intrinsic hand muscles with spasticity, wasting of the shoulder girdle, fasciculation of upper-extremity muscles, decreased sensation, dysesthesias, loss of vibration sense, hyperreflexia, clonus, upgoing toes, positive Hoffmann sign, and difficulty with heel, toe, and tandem walking and standing in Romberg position with eyes closed.  The imaging modality of choice in identifying the anatomical lesion is MRI.  Although more work with regard to the sensitivity and specificity of MRI in the detection of disk herniation has been done in the lumbar spine, the usefulness of this tool is widely recognized. [28, 29] If diagnostic uncertainty exists after history, examination and imaging, somatosensory-evoked potential examination can show the presence of neuronal dysfunction.  In patients with the acute development of symptoms suggestive of CM, as in the case presented here, it is important to consider vascular disorder such as spinal cord arteriovenous malformation as well as nonmyelopathy acute disorders such as Guillain-Barré syndrome or shoulder amyotrophy. 
This case is interesting because the patient was scheduled to receive conservative treatment that included cervical manipulation, but just before receiving treatment, his condition progressed to CM. Had this natural progression taken place after the patient received manipulation, the progression could have been improperly attributed to the manipulation rather than to natural history of the condition. Senstad et al  reported 6 cases in which patients experienced catastrophic episodes (2 strokes, 1 myocardial infarction, 1 bleeding aneurism, 1 upper extremity paresis, and 1 cauda equina syndrome) just before receiving manipulation or in circumstances in which they attempted to pursue manipulative care but did not receive such care. Had manipulative treatment been provided in these cases, the treatment could have been implicated in the incident, when it evidently would have occurred anyway.
In the case reported here, the patient was initially diagnosed with cervical radiculopathy secondary to disk herniation. Myelopathy was ruled out based on history and examination. Painful joint dysfunction was found at the level and on the side of the disk herniation. This led to the decision to include cervical manipulation in the treatment plan. Between the time of the examination and the first treatment, the patient developed CM.
This report is particularly important in light of recent publications reporting cases that were described as postmanipulative complications. Malone et al,  in a retrospective chart review from a 6-doctor neurosurgical practice, reported 22 patients whom they described as having had a significant complication to cervical manipulation. Of these, 21 had radiculopathy, 11 CM, 2 Brown-Sequard syndrome, and 1 vertebral artery occlusion secondary to far lateral disk herniation. Oppenheim et al  reported 18 patients who had cervical manipulation and who developed neurologic symptoms some time after treatment. This article led to a commentary  and lively debate via letters to the editor. In each of these case series, no definitive conclusions can be drawn regarding a cause-effect relationship between the manipulation and the radiculopathy or CM. As can be seen in the case reported here, disk herniation can progress on its own to CM without provocation. Were the patient reported here treated with manipulation on the first visit, this case could have been considered to be a “complication to manipulation” because of progression to CM after the manipulative treatment. Because the progression occurred even in the absence of manipulation, this shows that the progression would have occurred independent of the treatment.
Murphy et al  reported on 27 consecutive patients with imaging documentation of spinal cord compression from disk herniation or degenerative changes who were treated with manipulation for cervical joint dysfunction. In each case, no complications relative to the development of CM were noted. Although it is plausible that progression of cervical disk herniation to CM may be precipitated by manipulation in certain isolated cases, this case series argues against the likelihood that this is a common occurrence. Nonetheless, it stands to reason that practitioners of manipulation should take particular care when treating patients who have known pathology that has the potential to progress to CM. It is also important for spine practitioners to recognize CM when it occurs.
This case illustrates the importance of careful interpretation of posttreatment complications of any kind, in light of the natural history of the condition under consideration. Progression of neck and/or arm pain to myelopathy that occurs after manipulative treatment cannot be automatically assumed to relate directly to the manipulation because the progression may simply be a result of the natural history of the condition. It is important for authors of reports of myelopathy-related “complications to cervical manipulation,” and editors reviewing such reports, to consider this in the reporting of these complications.