The December 2002 issue of Neurosurgical Focus (an Internet publication of the Journal of Neurosurgery and the American Association of Neurological Surgeons1) contains an article authored by a group of neurosurgeons from Tulsa, Oklahoma, that suggests the safety of cervical spinal manipulation therapy (CSMT) has been overstated in the literature.2 The authors conclude, "Cervical spinal manipulation therapy may worsen preexisting cervical disc herniation or cause disc herniation resulting in radiculopathy, myelopathy, or VA compression. In cases of cervical spondylosis, CSMT may also worsen preexisting myelopathy or radiculopathy. Manipulation of the cervical spine may also be associated with higher complication rates than previously reported."
The study population was made up entirely of patients who received CSMT and were subsequently treated in a group neurosurgical practice during the five-year period ranging from Jan. 1, 1993, to Jan. 1, 1998. In total, 172 patients received CSMT, 32 of whom reported aggravation of symptoms following manipulation. A subset of 22 patients was considered markedly worse during, or immediately after, CSMT, and developed "irreversible" CSMT-related complications. The paper focused on the series of 22 patients negatively affected by CSMT. The authors indicated that the purpose of the study was to report the types of complications documented in a single-group practice of six neurosurgeons. They also attempted to estimate the actual regional incidence of such complications, based on these 22 cases.
The study was carried out by retrospectively reviewing the medical records of all patients who were treated by manual practitioners or underwent spinal manipulation prior to presenting to the neurosurgeon practices. The greatest portion of the paper provided a description of each of the selected cases relative to the basic history of CSMT involvement; the sequelae that followed CSMT; the surgical procedure provided; and the eventual patient outcome.
In reviewing this paper, it became apparent that both its methodology and conclusions were questionable.
Methodologically, this study was based solely on Class III evidence, from which no cause-and-effect conclusions can be drawn. The study was merely a collection of case reports, which may be somewhat interesting, but cannot be used to make inferences related to cause.
A second methodological concern is the likelihood for patient recall bias. It is possible that the fundamental beliefs and perspectives of the patients involved may have influenced their recollection of events. This issue is important, since the researchers relied heavily upon patient recall of events surrounding their CSMT histories. These patients may have been subject to the influence of recall bias by laying blame on the CSMT practitioners for their poor outcomes. As a result, they may have been inclined to exaggerate certain aspects of the care they received.
A third methodological concern is the likelihood of bias on the part of the neurosurgeons. The doctors involved in this study, like other neurosurgeons, are the principal "end of the line" providers with regard to most poor neurological outcomes. As such, they routinely see the worst-case scenarios of CSMT providers. This nonrepresentative patient exposure may have produced a prejudiced viewpoint relating to CSMT procedures, and CSMT practitioners in general. This becomes apparent when one compares the findings of this study with those of other researchers who have reported far fewer severe reactions related to CSMT. One notable study concerning unpleasant reactions following chiropractic manipulation concerned 4,712 visits to chiropractors in Norway. It was discovered that transient, mild-to-moderate symptoms were common, but there were no reports of serious complications in any patients.3
A fourth methodological concern is that the researchers did not account for the natural course of the patient's condition that may have taken place absent CSMT. A more valid means of assessing the stated objectives of the study would have been to compare the types and frequency of complications between two large, randomly assigned groups of patients. One group would undergo CSMT following certain protocols, while the other group would receive a sham treatment. From such a study, negative and positive outcomes could be more accurately assessed. Without this type of design, however, there is no way to be confident that CSMT actually influenced natural progression of the disease.
A fifth methodological concern is that there was no consistent delineation of the interval between the CSMT and the intensification of symptoms. When the interval was specified, there was no statement regarding any independent effort to verify its accuracy. If a patient initially reported that the onset of intensification immediately followed the CSMT, but independent verification found the duration was really hours or days, the causation issue becomes more obscure. Natural progression aside, one's confidence in ascribing causation wanes as duration increases.
With such significant concerns over the methodology of this study, its conclusions are inherently suspect. Let's examine each of the previously stated conclusions sequentially.
"Cervical spinal manipulation therapy may worsen preexisting cervical disc herniation or cause disc herniation resulting in radiculopathy, myelopathy or VA compression." This conclusion apparently was based on their review of the 22 cases presented. Such Class III evidence is insufficient to determine causal relationships. Furthermore, cervical disc herniations often are characterized by a history of gradual onset, with the first sign of cervical disc herniation often merely a restriction of neck extension. The symptomatic picture typically is not fully developed in the beginning, and the patient may present with what appears to be a cervical sprain or stiff neck. Quite often, orthopedic and neurological tests do not reveal radicular signs and symptoms on initial presentation. Muscle weakness, depressed reflexes and other neurologic signs may not appear for days or weeks after the relatively benign initial symptoms.4,5 Frequently, this condition resolves with time and conservative treatment (including CSMT). Surgery is indicated only when appropriate conservative treatment, administered over a reasonable period of time, has failed.6 Nevertheless, this condition occasionally follows an unfavorable course, leading to deterioration no matter what treatment regimen is established. If CSMT was applied during this period, blame may be assigned to the practitioner, when none actually exists. This may have been the case in some of the examples in the current study.
Some of the cases presented were extreme, with eight involving free fragments of extruded disc material, a condition known to be unstable and prone to deterioration. The fact that five of these cases experienced a rapid deterioration subsequent to CSMT may be related to this established instability. Extruded disc material is capable of shifting, and almost any moderate neck motion may move extruded material into a damaging position. Unfortunately, often it is not possible to distinguish these cases from more stable disc protrusions, or even common cervical strain, until more significant signs and symptoms have evolved.
The authors suggest that the issue regarding contraindications to high-velocity neck manipulation in the presence of varying grades of disc herniation has been resolved, but disagreement still exists.7 Perhaps the best assessment of the chiropractic profession's position on this issue can be found in the consensus opinion of The Mercy Guidelines. These guidelines specify that manipulation is absolutely contraindicated in the presence of "extreme extrusions with severe neurological deficit."8 Moreover, spinal manipulation and mobilization have been shown to be beneficial in cases of cervical disc herniation, and may restore normal ROM and decrease pain.9,10 A study conducted by BenEliyahu involved 27 patients with MRI-documented cervical or lumbar disc herniations who were treated with chiropractic manipulation. Care consisting of traction for the cervical spine or flexion distraction in the lumbar spine in the acute phase, and rotational manipulation in the subacute phase, had an 80% success rate, with no complications.11 Based on these findings, and a review of the literature, BenEliyahu indicated that manipulation was safe for the treatment of disc herniations. Additionally, he mentioned that rotational manipulation was not likely to result in disc failure or exacerbation of a disc herniation, since rotational forces associated with manipulation of a healthy disc would result in facet fracture before disc failure would occur.
"In cases of cervical spondylosis, CSMT may also worsen preexisting myelopathy or radiculopathy." Based on this statement, one wonders if the implication is that no patient with cervical spondylosis should undergo CSMT. There is no effort by the authors to quantify the degree or location of the spondylosis they are referring to in their conclusion. Cervical spondylosis, even in moderate to severe cases, is not an absolute contraindication to CSMT.8 Certainly, the 10 cases presented by these authors in which spondylosis involved complications were temporally associated with CSMT, are insufficient to justify this conclusion.
"Manipulation of the cervical spine may also be associated with higher complication rates than previously reported." Based on their limited and suspect data, the authors calculated the risk of developing an irreversible CSMT complication at 1:8,500 cervical manipulations. The average DC sees approximately 115 visits per week12 for 50 weeks during a typical year, or 5,750 total visits. Approximately 75 percent of chiropractic visits (4,312 total visits) entail cervical manipulationt. If the 1:8,500 complication rate was true, there would be one irreversible CSMT complication every two years - for every practicing DC in this country. This does not appear to be reasonable, and no data are available to support this notion. Anecdotally, the authors of this critique have a combined total of 50 years in practice, and during this time, there has only been one such case that would meet the study's criteria for an irreversible complication.
We agree that the complications inherent to any procedure need to be accurately researched and quantified. However, when the methods of data collection and analysis avoid the strict protocols of the scientific method, any resulting conclusions are of little value. At best, the study by Malone, et al., is little more than a poorly structured series of anecdotal cases.
References
- Neurosurgical Focus. Internet publication of the Journal of Neurosurgery and the American Association of Neurological Surgeons, 2003. www.neurosurgery.org/focus/.
- Malone DG, et al., Complications of cervical spine manipulation therapy: 5-year retrospective study in a single-group practice. Neurosurg Focus 2002:13(6).
- Senstad OC, Leboeuf-Yde C, Borchgrevink C. Frequency and characteristics of side effects of spinal manipulative therapy. Spine, 1997;22(4): p. 435-40; discussion 440-1.
- Turek SL, Weinstein SL, Buckwalter JA. Turek's Orthopaedics: Principles and their application. 5th ed. 1994, Philadelphia: Lippincott. cm.
- Wiesel SW, Delahay JN. Principles of Orthopaedic Medicine and Surgery. 2001, Philadelphia, Pa.: W.B. Saunders. viii, 859.
- Heckmann JG, et al. Herniated cervical intervertebral discs with radiculopathy: an outcome study of conservatively or surgically treated patients. J Spinal Disord, 1999;12(5): pp.396-401.
- Haneline M. Chiropractic management of cervical intervertebral disc herniation. Journal of the Neuro musculoskeletal System, 2001;9(1): p. 13-15.
- Haldeman, S, Chapman-Smith D, Petersen DM. Guidelines for chiropractic quality assurance and practice parameters: proceedings of the Mercy Center Consensus Conference. 1993, Gaithersburg, Md.: Aspen Publishers. xli, 222.
- Herzog J. Use of cervical spine manipulation under anesthesia for management of cervical disk herniation, cervical radiculopathy, and associated cervicogenic headache syndrome. J Manipulative Physiol Ther, 1999;22(3): p.166-70.
- Furman MB, Puttlitz KM. Cervical Disc Disease. eMedicine, 2001.
- BenEliyahu DJ. Magnetic resonance imaging and clinical follow-up: study of 27 patients receiving chiropractic care for cervical and lumbar disc herniations. J Manipulative Physiol Ther, 1996;19(9): p. 597-606.
- Goertz C. ACA Summary of the 2000 ACA professional survey on chiropractic practice. Journal of the American Chiropractic Association, 2001;38(2): p. 27-30.
Michael T. Haneline, DC, FICR
El Cajon, California
Gary Lewkovich, DC
San Marcos, California
