Clinical/Scientific Notes
Vertebral artery dissection and stroke
following neck manipulation by Native
American healer
J. George Quintana, MD; Eric C. Drew, MD;
Thomas E. Richtsmeier, MD; and Larry E. Davis, MD
Stroke as a complication of cervical manipulation and dissection of
the vertebral artery is rare but has been well described1 and is
most commonly associated with chiropractic manipulation of the
neck using a technique often called “high velocity thrust.”2 Strokes
from vertebral artery dissections have also been reported with
cervical self-manipulation and other causes of neck trauma such
as sports and roller coasters. The authors report on a patient with
vertebral artery dissection and a stroke occurring after a Native
American healing ceremony.
Case report. This previously healthy 32-year-old woman from
the Pueblo of Zuni presented with the sudden onset of vertigo,
left-sided ataxia, and vomiting. Six days before the onset of her
symptoms, the patient underwent a healing ceremony with a native healer whom she referred to as a “bone setter” for complaints
of nontraumatic left shoulder discomfort. During the ceremony,
she underwent vigorous manipulation of her neck with a sudden
thrusting of the head upward and to the right while her left
shoulder was being held to the ground. She noted neck discomfort
immediately afterward, which persisted. She awoke on day 6 vertiginous with left-sided ataxia. She had no identifiable risk factors
for stroke or vertebral artery dissection and no family history of
cerebrovascular disease.
She had normal mentation, cranial nerves, sensation, strength,
and physiologic deep tendon reflexes with slow inaccurate responses on left finger-to-nose testing and on rapid alternating
movements. Her gait was wide based, and she leaned to the left.
The results of routine laboratory studies, a rapid plasma reagin
test, and prothrombin time were normal.
MRI showed acute infarction in the midline left cerebellar
hemisphere and vermis in the distribution of the left posterior
inferior cerebellar artery (figure, A). MR angiography demonstrated likely occlusion of the left vertebral artery. On vertebral
arteriography, severe narrowing of the left vertebral artery was
identified beginning at the level of the C2 vertebral body and
extending upward (see the figure, B). At the level of C1, collateral
blood flow was seen entering the left vertebral artery from a small
artery that arose from a branch of the thyrocervical trunk. No
blood flow was seen in the left posterior inferior cerebellar artery
(PICA).
A diagnosis was made of left vertebral artery dissection with a
probable embolus originating from the distal vertebral artery
stump and occluding the PICA. The patient improved and was
discharged on warfarin anticoagulation therapy. At 6 months, she
had a normal gait with minimal abnormalities on left finger-tonose testing. Repeat MRI and MR angiography were unchanged.

The warfarin was discontinued without subsequent neurologic
events for 1 year.
Discussion. This case illustrates the occurrence of PICA infarction and vertebral artery dissection as sequelae of vigorous cervical manipulation performed by a Native American healer during a
healing ceremony. The patient’s signs and symptoms were typical
for a vertebral artery dissection. Neck pain develops in one-half of
such patients, and headaches occur in two-thirds, often beginning
within hours of the neck manipulation, although cerebral ischemic
signs may be delayed for days to 2 weeks.1 The outcome of the
stroke depends on the size and severity of the initial infarction
and the extent of collateral circulation. In our patient, a small
collateral artery appeared to carry blood around the dissection
(see the figure, B), which may have contributed to her good outcome. Overall, about 75% of patients make a good recovery, but
some patients are permanently quadriplegic from a brainstem
infarction, and approximately 5% die.1
Most Native American healing ceremonies do not involve vigorous neck manipulations. Vigorous neck and head manipulation
apparently can occur in healing ceremonies conducted by a “bone
setter,” however. Bone setters are one of 12 Zuni medicine fraternities.3 Tradition has it that a bone setter must be a survivor of a
lightning bolt strike who then gains special powers to heal
bones.4,5 These individuals apprentice with established bone setters for variable periods of time so as to learn traditional methods
of healing diseases of bones and joints. Because Zuni healing
ceremonies are rarely observed by physicians, it is not known how
this neck manipulation compares with chiropractic manipulation.
Worldwide, many indigenous and ethnic groups, including Native Americans, have a long history of traditional medical healers
called bone setters who treat bone and joint problems.6,7 Complications from their nonmedical approach to healing bones are recognized.6,7 It is not known how many bone setters practice vigorous
cervical manipulation and thus are at risk of causing stroke from
traumatic dissection of a vertebral artery.
Acknowledgment
The authors thank Dr. Theodore R. Frisbie, Professor Emeritus, Department of Anthropology, College of Arts and Sciences, Southern Illinois University at Edwardsville, IL, and Ms. Mary Alice Tsosie, University of New
Mexico, Center for Southwest Research, Zimmerman Library, Albuquerque,
NM, for their help in the research on bone setters.
From the Neurology Service (Drs. Quintana, Drew, and Davis), New Mexico
VA Health Care System, Albuquerque; Department of Neurology (Drs. Quintana, Drew, and Davis), University of New Mexico School of Medicine,
Albuquerque; and Department of Internal Medicine (Dr. Richtsmeier), Gallup Indian Medical Center, Gallup, NM.
Received November 5, 2000. Accepted in final form January 15, 2002.
Address correspondence and reprint request to Dr. Larry E. Davis, Neurology Service, New Mexico VA Health Care System, 1501 San Pedro Drive SE,
Albuquerque, NM; e-mail: LEDavis@unm.edu

Figure. (A) Intermediate-weighted MRI scan 2
days after the onset of stroke showing a cerebellar
stroke in the midline left cerebellar hemisphere and
vermis in the territory of the left posterior cerebellar
artery (arrow). (B) Left vertebral arteriogram with
subtraction 2 days after the onset of stroke demonstrating severe narrowing of the artery beginning at
the level of the C2 vertebral body and extending
upward to the level of C1 (arrow). Note a collateral
artery, likely a branch of the thyrocervical trunk,
entering the vertebral artery above the level of the
stenosis at about the level of the first cervical vertebra (arrowhead).
1434

NEUROLOGY 58

May (1 of 2) 2002

Copyright © 2002 by AAN Enterprises, Inc.

References
1. Schievink WI. Spontaneous dissection of the carotid and vertebral arteries. N Engl J Med 2001;344:898 –906.
2. Vickers A, Zollman C. ABC of complementary medicine: the manipulative therapies: osteopathy and chiropractic. BMJ 1999;319:1176 –1179.
3. Lyon WS. Encyclopedia of Native American healing. Santa Barbara, CA:
ABC-CLIO, 1996:313.

Absence of elevated anti–␣-synuclein and antiEBV latent membrane protein antibodies
in PD
J.M. Woulfe, MD, PhD; R. Duke, MD; J.M. Middeldorp, PhD;
S. Stevens, MSc; M. Vervoort, BSc; M. Hashimoto, MD, PhD;
E. Masliah, MD; P. Chan, MD, PhD; D.A. Di Monte, MD;
J.W. Langston, MD; G. Petzinger, MD;
H. Hoogendoorn, MLT; and D.G. Munoz, MD
We recently demonstrated that commercially available monoclonal antibodies generated against the latent membrane protein 1
(LMP1) of Epstein–Barr virus (EBV) cross-react with the neuronal
protein ␣-synuclein.1 LMP1 is a virally encoded protein expressed
during EBV latency. Up to 22% of healthy EBV-seropositive individuals generate a humoral immune response against this protein
as part of a lifelong immune surveillance of EBV.2,3 Whether this
cross-reaction has any biological relevance remains to be determined.
There is growing evidence for a primary role of ␣-synuclein aggregation in the pathogenesis of PD.4 Insoluble ␣-synuclein filaments accumulate in brains of patients with PD. Recombinant ␣-synuclein
protein can assemble in elongated filaments with ultrastructural
features similar to those of Lewy body filaments observed in situ. In
light of this as well as the fact that EBV latently infects approximately 90% of the human population worldwide and represents the
target of a lifelong immune response, we were interested in exploring
the possibility that molecular mimicry between EBV (and specifically
the LMP1 protein) and ␣-synuclein could be involved in the pathogenesis or progression of PD.
Methods and results. In the first phase of the investigation, we
performed EBV-specific ELISA on serum samples from 10 patients
with PD and 10 age-matched controls. All 20 samples showed evidence of latent EBV infection with detectable titers of antibodies
against viral capsid antigen (VCA)-p18 and the latent nuclear protein Epstein–Barr nuclear antigen (EBNA)-1. For the detection of
anti-LMP1, purified LMP1 protein was coated onto ELISA plates,
separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis, and transferred to nitrocellulose in a parallel immunoblot
analysis. Both assays are highly reproducible and have been described previously.3 Only three samples (two controls and one PD)
displayed detectable levels of antibodies against LMP1.
We then analyzed serum samples from 28 patients with PD
and 19 control subjects for antibodies against ␣-synuclein using a
two-step sandwich ELISA. Briefly, the ELISA was performed as
described previously1 with the following modifications. A reference
standard was prepared using a monoclonal antibody to human
␣-synuclein (Zymed, San Francisco, CA) added to human plasma.
Reference plasma and test samples were diluted at least 25-fold
before application to the plate. A peroxidase-labeled antibody solution reactive to both human and murine IgG was used as the
detecting antibody. The results were expressed as monoclonal antibody equivalents in micrograms per milliliter. All subjects
showed low levels of serum anti-synuclein antibodies (0.055 to
0.276 ␮g/mL), with the exception of three from each group who
showed higher anti-synuclein antibody levels (0.301 to 0.795 ␮g/
mL). Notably, there was no difference in the mean anti-synuclein
antibody reactivity between PD (0.178 ␮g/mL) and control (0.216
␮g/mL) groups (p ⫽ 0.37).
In light of the negative results we obtained based on serum analysis and to address the possibility of CNS localization of an EBV
infection and immune response, we applied the same sandwich
ELISA to CSF samples from five PD cases and four controls. All
samples showed extremely low to undetectable levels of antisynuclein antibodies (0.006 to 0.015 ␮g/mL). There was no difference
in the mean values between PD and control samples (p ⫽ 0.28).
Finally, we used a highly sensitive PCR technique for the de-

4. Benedict R. The husband who was pregnant. In: Zuni mythology. New
York: Columbia University Press, 1935:107–109.
5. Parsons EC. Pueblo Indian religion. Lincoln, NE: University of Nebraska
Press, 1939:427.
6. Onuminya JE, Obekpa PO, Ihezue HC, Ukegbu ND, Onabowale BO.
Major amputations in Nigeria: a plea to educate traditional bone setters.
Trop Doct 2000;30:133–135.
7. Hatipoglu S, Tatar K. The strengths and weaknesses of Turkish bonesetters. World Health Forum 1995;16:203–205.

tection of EBV genome5 to address the possibility of local CNS
infection. None of the PD or control CSF samples contained detectable EBV DNA sequences. In addition, none of 17 formalinfixed and paraffin-embedded tissue sections, including four
samples of PD midbrain, three samples of PD cortex, four samples
of cortex from dementia with Lewy bodies, three samples of cortex
from controls, and three samples of midbrain from controls, showed
detectable EBV DNA sequences using either a single-copy EBNA-1
PCR or a BamH1-W repeat PCR with a sensitivity of 1 to 2 EBV
genome–positive cells per assay. These PCR-negative samples became positive after spiking with EBV plasmid DNA, indicating absence of PCR inhibition and true EBV negativity of the samples.
Moreover, all samples were positive using the same technique for the
detection of ␤-globin DNA, indicating good DNA quality.
Discussion. The results of these studies do not provide evidence that an aberrant systemic immune response to latent EBV
infection resulting in molecular mimicry between EBV LMP and
␣-synuclein has a role in the pathogenesis of PD. In addition, in
our parallel analysis of a small group of CSF samples from PD
patients and controls, we failed to reveal evidence for elevated
immune reactivity to synuclein or LMP1 in PD generated exclusively within the CNS. Although the CSF results are consistent
with the serum results, the sample size is small, and further
studies of CSF in a larger cohort are needed to definitively exclude
a discrepancy in anti-synuclein and anti-LMP1 antibody titers
among PD and control subjects. Indeed, the mere presence of serum
anti-synuclein antibodies (albeit at low levels) demonstrated in our
studies is intriguing. Specifically, the possibility that low levels of
circulating anti-synuclein autoantibodies may serve a protective role
with respect to ␣-synuclein aggregation should be explored. This is
particularly relevant in light of recent studies demonstrating that
naturally occurring antibodies directed against ␤-amyloid exist in
human CSF and plasma and that titers of these antibodies are reduced in AD.6 This finding is provocative in light of recent studies
showing immune-mediated prevention and even reversal of
Alzheimer-type changes secondary to immunization with the
␤-amyloid peptide in a mouse model of AD.7
From the Department of Pathology (Dr. Woulfe), The University of Ottawa
and Department of Laboratory Medicine, The Ottawa Hospital, and the
Department of Medicine (Neurology) (Dr. Duke), Hamilton Health Sciences
Center, Hamilton, Ontario, Canada; Department of Pathology (Dr. Middeldorp, S. Stevens, and M. Vervoort), Free University Hospital, Amsterdam,
the Netherlands; Department of Neurosciences (Drs. Hashimoto and
Masliah), University of California at San Diego, School of Medicine, La
Jolla; The Parkinson’s Institute (Drs. Di Monte, Langston, and Petzinger),
Sunnyvale, CA; Affinity Biologicals (H. Hoogendoorn), Hamilton Civic Hospitals Research Centre, Hamilton, and Departments of Pathology and Clinical Neurological Sciences (Dr. Munoz), University of Western Ontario,
London, Ontario, Canada; and Servicio de Neurologia (Dr. Munoz), “Doce de
Octubre” Hospital, Madrid, Spain.
Supported by The National Parkinson Foundation, Inc., Miami, FL.
Received March 19, 2001. Accepted in final form January 20, 2002.
Address correspondence and reprint requests to Dr. John Woulfe, Department of
Pathology, The Ottawa Hospital, Civic Campus, 1053 Carling Avenue, Ottawa,
Ontario, Canada K1Y 4E9; e-mail: jwoulfe@ottawahospital.on.ca
Copyright © 2002 by AAN Enterprises, Inc.

References
1. Woulfe J, Hoogendoorn H, Tarnopolsky M, Munoz DG. Monoclonal antibodies against Epstein–Barr virus cross-react with ␣-synuclein in human
brain. Neurology 2000;55:1398 –1401.
2. Rowe M, Fink J, Szigeti R, Klein G. Characterization of the serological
response in man to the latent membrane protein and the six nuclear
antigens encoded by Epstein–Barr virus. J Gen Virol 1988;69:1217–1228.
3. Meij P, Vervoort MBHJ, Aarbiou J, et al. Restricted low-level human
May (1 of 2) 2002

NEUROLOGY 58

1435

Vertebral artery dissection and stroke following neck manipulation by Native American
healer
J. George Quintana, Eric C. Drew, Thomas E. Richtsmeier, et al.
Neurology 2002;58;1434
DOI 10.1212/WNL.58.9.1434
This information is current as of May 14, 2002
Updated Information &
Services

including high resolution figures, can be found at:
http://n.neurology.org/content/58/9/1434.full.html

References

This article cites 4 articles, 1 of which you can access for free at:
http://n.neurology.org/content/58/9/1434.full.html##ref-list-1

Subspecialty Collections

This article, along with others on similar topics, appears in the
following collection(s):
Infarction
http://n.neurology.org//cgi/collection/infarction

Permissions & Licensing

Information about reproducing this article in parts (figures,tables) or in
its entirety can be found online at:
http://n.neurology.org/misc/about.xhtml#permissions

Reprints

Information about ordering reprints can be found online:
http://n.neurology.org/misc/addir.xhtml#reprintsus

Neurology ® is the official journal of the American Academy of Neurology. Published continuously since
1951, it is now a weekly with 48 issues per year. Copyright . All rights reserved. Print ISSN: 0028-3878.
Online ISSN: 1526-632X.