Reflection and Reaction
Alter J, Lou F, RabinowitzA, etal. Systemic delivery of morpholino Hoffman EP. Skipping toward personalized molecular medicine.
oligonucleotide restores dystrophin expression bodywide and improves N EnglJ A/led 2007; 357: 2719-22.
dystrophic pathology. Nat A/led 2006; 12:175-77- Aartsma-RusA, Fokkema IF, VerschuurenJJ, etal. Theoretic applicability of
Mann CJ, Honeyman K, Cheng AJ, et al. Antisense-induced exon skipping antisense-mediated exon skipping for Duchenne muscular dystrophy
and synthesis of dystrophin in the mdx mouse. ProcNat/AcadSdUSA 2001; mutations. Hum Mutat 2009; 30: 293-99.
98: 42-47 Aartsma-Rus A, Janson AA, Kaman WE, et al. Antisense-induced multiexon
van DeutekomJC, Bremmer-Bout MJanson AA, etal. Antisense-induced skipping for Duchenne muscular dystrophy makes more sense.
exon skipping restores dystrophin expression in DMD patient derived AmJ Hum Genet 2004; 74: 83-92.
muscle cells. Hum Mol Genet 2001; 10:1547-54. YokotaT, LuQL, PartridgeT, etal. Efficacy of systemic morpholino
van DeutekomJC, Janson AA, Ginjaar IB, et al. Local dystrophin exon-skipping in Duchenne dystrophy dogs. Ann Neurol 2009; 65: 667-76.
restoration with antisense oligonucleotide PROO5I. N EnglJ A/led 2007; 13 van Vliet L, de Winter CL, van Deutekom JC, van Ommen GJ, Aartsma-Rus A.
357: 2677-86. Assessment of the feasibility of exon 45-55 multiexon skipping for
Kinali M, Arechavala-GomezaV, Feng L, etal. Local restoration of Duchenne muscular dystrophy. BMCMed Genet 2008; 9:105.
dystrophin expression with the morpholino oligomerAVI-4658 in 14 Aartsma-Rus A, Kaman WE, Wei] R, den Dünnen JT, van Ommen GJ, van
Duchenne muscular dystrophy: a single-blind, placebo-controlled, dose- DeutekomJC. Exploring the frontiers of therapeutic exon skipping for
escalation proof-of-concept study. Lancet Neurol 2009; published online Duchenne muscular dystrophy by double targeting within one or multiple
August 26. DOI: 10.10l6/S1474-4422(09)70211-X. exons. A/lolTher 2006; 14:401-07
Heemskerk HA, DeWinterCL, de KimpeSJ, etal. In v/vo comparison of
2'-0-methyl phosphorothioate and morpholino antisense oligonucleotides for
Duchenne muscular dystrophy exon skipping.] Gene A/led 2009; 11:257-66.
Cervicogenic headache: a pain in the neck for some
In this issue of The Lancet Neurology, Bogduk and Govind of the C3 afférents; this is certainly consistent with the see Review page 959
review the vexed topic of cervicogenic headache.1 As a laboratory anatomy.8 What would the convergence of
neurologist interested in headache, it seems self-evident these afférents predict? The data suggest that nociceptive
that this topic should be of interest. Much of what has activation in structures innervated by either trigeminal
passed for science in this field is rightly criticised by the or upper cervical afférents might result in a perception of
authors, and there seems no benefit in trawling over the pain that is not anatomically related to the structure with
arguments, for ifthere was no controversythere would be the pathology. Put simply, the anatomy predicts that the
no issue. As the authors say, the anatomy and physiology simple clinical localisation process of "pain marks the
are clear enough,2 so what is the problem? There are spot" is doomed to fail. Moreover, the data all suggest
several issues, again largely and capably discussed by that cervicogenic headache—headache from activation
Bogduk and Govind; I would like to highlight some of of nociceptors in the upper cervical spine—should exist.
these from the neurologist's perspective, as I think we An important aspect of the apparent controversy in this
need to get our house in order. The three particular topics field might relate substantially to referral bias. In the USA,
of note are: the clinical implications of the anatomy and more than 90% of patients who presentto a physician with
physiology, the problem of referral bias, and the issue of disabling headache lasting longer than 3 months have
primary versus secondary headache from a neurologist's migraine.9 This predominance of migraine is likely to feed
point of view. Finally, how should we proceed? into neurology practice, which means that, whatever the
Perhaps the most basic issue revolves around the prevalence of cervicogenic headache, neurological practice
anatomy and physiology of upper cervical segment will be swamped with cases of migraine. If the above
nociceptive afférents and their projections to second- analysis of the anatomy and physiology is accepted, and
order neurons. In experimental animals, the cervical the data on neck symptoms in patients with premonitory
and ophthalmic division of trigeminal neurons clearly symptoms suggest that about 90% of these patients have
synapse on common second-order neurons34 in the these symptoms,10 then most patients seen in neurology
trigeminocervical complex.5The basic data are supported or headache practice with neck discomfort have migraine.
by clinical observations, such as referred pain from This assumption does not mean cervicogenic headache
cervical muscles6 and the C2-3 zygapophysial joint.7 does not exist, it does not demean it, nor imply anything
Interestingly, the clinical data reviewed1 suggest that the negative; this is just referral bias. In a recent retrospective
caudal limit of cranial referral in the neck is at the level study of patients who responded to occipital nerve
www.thelancet.com/neurology Vol 8 October 2009 875
Reflection and Reaction
stimulation (importantly, done in collaboration with even controlled blocks can only be part of the analysis—
a pain specialist), half the responders had headache of without the detailed clinical picture, these blocks alone
cervical origin and about half had migraine;11 yet, in my will not advance our understanding.
clinical practice, not one patient with headache of cervical Finally, as most practitioners do not have facilities
origin has been implanted with an occipital stimulator out for procedures such as diagnostic blocks under X-ray
of more than 50 patients. Why would this be? Is it that control and, importantly, as the procedures do not in
patients with neck trauma that produces pain are just sent isolation offer a definitive biological outcome, how do
to pain doctors, who do a very good job with them? When we proceed? The term "cervicogenic headache" was
patients with migraine are sent to pain doctors, and local introduced by Sjaastad and colleagues in 198313 with
measures fail, the patients find their way to neurology, revised criteria proposed in 199014 and 1998.15The
thus giving neurologiststhetwo-prongedfalse impression International Headache Society provides diagnostic
that true cervicogenic headache does not exist. It is almost criteria that, unfortunately, overlaps in two places in
self-evident that cervicogenic headache exists: the neck its classification: sections 11.2.1 and 13.12.16 It would
has the necessary structures and trauma is conceivable. be ideal, indeed cost-effective, if we could diagnose
The problem is how common is this type of headache and cervicogenic headache by history. Unfortunately, and key
how do we spot it? The fact that cervicogenic headache to the hypothesis advanced by Bogduk and Govind (and
might not be common in neurology clinics should not indeed consistent with the anatomy and physiology),
confound either its understanding or management, as history-taking will most often not suffice. The only
this strategy will not help patients. way forward is collaboration to test clinical history and
An important issue, which is very complex to unravel, examination, underpinned by local procedures. For
is the intersection of neck pathology with a primary example, my impression is that caudad pressure on the
headache, typically migraine. If neck pathology in a head with lateral rotation is a very sensitive marker of
genetically susceptible individual produces, triggers, upper cervical pathology (the cervical grinding test),
or worsens the underlying predisposition to primary but such clinical tests need the careful approach that
headache, what should we call it? If the previous Bogduk has pioneered. Moreover, unless neurology and
headache type is made worse in close temporal proximity pain medicine can work together on prospective studies
to a neck injury than the International Headache Society of diagnosis and treatment, this question will be as
classification says, judgment is required! If a patient controversial in two decades from now as it has been in
has migraine that is normally triggered by red wine and the past two decades. As this lack of consensus would not
drinks red wine 16 times a month, and has headache each help our patients, non-collaboration is just unacceptable.
time 6 h later, they have migraine rather than chronic
daily headache of the red wine type. One could take the Peter] Goadsby
same view of migraine triggered by skipping meals, say Headache Group-Department of Neurology, University of
California, San Francisco, 1635 Divisadero Street, San Francisco,
17 days a month, or even, as I have seen, a daily afternoon
CA 94115, CA, USA
headache of 4-5 h duration triggered by a morning dose firstname.lastname@example.org
of glyceryl trinitrate. Each case is migraine, with a very I have consulted for and done research funded by neuromodulation companies
distinct trigger. So, if a patient is migrainous (ie, has the (Boston Scientific and Medtronic) interested in occipital nerve stimulation,
although this work has not been in the sphere of cervicogenic headache.
genetic predisposition), and has headache aggravated by
1 Bogduk N, Govind J. Cervicogenic headache: assessing the evidence for clinical
a neck injury, then it is migraine. This does not judge the diagnosis, invasive tests, and treatment. Lancet Neural 2009; 8: 959-68.
treatment, nor does the treatment alter the underlying 2 Bartsch T, Goadsby PJ. Anatomy and physiology of pain referral in primary
and cervicogenic headache disorders. Headache Currents 2005; 2:42-48.
biology. The treatment of daily migraine induced by 3 Bartsch T, Goadsby PJ. Stimulation ofthe greater occipital nerve induces
increased central excitability of durai afferent input. Brain 2002;
glyceryl trinitrate is to stop glyceryl trinitrate, if possible;
the treatment of daily migraine triggered by neck 4 BartschT, Goadsby PJ. Increased responses intrigeminocervical nociceptive
neurones to cervical input after stimulation ofthe dura mater.Brain 2003;
pathology is to alleviate that pathology. Greater occipital 126:18OI-I3.
nerve injections with lidocaine and methylprednisolone 5 Goadsby PJ, Bartsch T. On the functional neuroanatomyof neck pain.
Cephalalgia 2008; 28 (suppl 1): 1-7
can be useful in migraine,12 but in no way imply that 6 Feinstein B, LangtonJNK, Jameson RM, Schiller F. Experiments on pain
there is C2 branch pathology in these patients. Thus, referred from deep somatic tissues. J ßonejo/nt Surgery 1954; 36A: 981-97
876 www.thelancet.com/neurology Vol 8 October 2009
Reflection and Reaction
Dwyer A, Aprill C, Bogduk N. Cervical zygapophysial joint pain patterns I: 12 Afridi SK, Shields KG, Bhola R, Goadsby PJ. Greater occipital nerve injection
a study in normal volunteers. Spine 1990; 15: 453-57 in primary headache syndromes—prolonged effects from a single injection.
Goadsby PJ, Hoskin KL.The distribution oftrigeminovascular afférents in Pain 2OO6; 122:126-29.
the nonhuman primate brain Macaca nemestrina: a c-fos 13 Sjaastad 0, Saunte C, Hovdal H, Breivik H, Greenback E. 'Cervicogenic'
immunocytochemical study.J Anatomy 1997; 190: 367-75. headache, an hypothesis. Cephala/g/a 1983; 3: 249-56.
Tepper SJ, Dahlof CG, Dowson A, etal. Prevalence and diagnosis of 14 Sjaastad 0, FredriksenT, PfaffenrathV. Cervicogenic headache: diagnostic
migraine in patients consulting their physician with a complaint of criteria. Headache 1990; 30:725-26.
headache: data from the landmark study. Headache 2004; 44: 856-64. 15 Sjaastad 0, FredriksenT, PfaffenrathV; the Cervicogenic Headache
Giffin NJ, Ruggiero L, Lipton RB, etal. Premonitory symptoms in migraine: International Study Group. Cervicogenic headache: diagnostic criteria.
an electronic diary study. Neurology 2003; 60: 935-40. Headache 1998; 38:442-45.
Paemeleire K, Van BuytenJP, Van Buynder M, etal. Phenotype of patients l6 Headache Classification Committee of the International Headache Society.
responsive to suboccipital neurostimulation for refractory head pain. The International Classification of Headache Disorders: 2nd edition.
EurJ Neurol 2008; 15 (suppl 3): 10. Cepfia/alg/a 2004; 24 (suppl 1): I-I60.
Can the WATCHMAN device truly PROTECT from stroke in
Atrial fibrillation causes 15-20% of ischaemic strokes device (via a catheter-based delivery system through
and the overall risk of stroke in patients with non- a trans-septal puncture) was compared with that of
valvular atrial fibrillation is about 5% per year globally.1 long-term warfarin therapy. More than 700 patients
Warfarin has long been the cornerstone for decreasing fulfilled the inclusion criteria of having non-valvular
risk of stroke in patients with atrial fibrillation and its atrial fibrillation, being suitable for anticoagulation, and
efficacy has been well established.1 However, 14-44% of having a CHADS2 risk score of 1 or more. Most of the
patients with atrial fibrillation who are at risk of stroke patients who had the device implanted (349 of 408)
are ineligible for anticoagulation therapy, mostly owing stopped warfarin at 45 days (as predefined) once there
to the risks of major bleeding and falls.2 Even in patients was transoesophageal echocardiographic confirmation
who are eligible, the risk of bleeding, inconvenience of LAA closure, and then remained on aspirin and
of frequent monitoring and dose adjustments, drug clopidogrel for 6 months after randomisation, followed
Interactions, and restrictions on diet and alcohol intake by long-term aspirin monotherapy. The probability
perhaps explain why warfarin discontinuation rates are of non-inferiority of the device was greater than
estimated to be as high as 38% per year.2 99-9% with regard to the primary efficacy outcome
More than 90% of atrial thrombi originate from the (occurrence of all types of stroke, cardiovascular or
left atrial appendage (LAA),3 and devices that can isolate unexplained death, or systemic emboli within up to
this structure from the systemic circulation and perhaps 3 years), and patients who received the device had
obviate the need for long-term anticoagulation therapy fewer haemorrhagic strokes than the controls. However,
have been developed. Despite early interesting and the primary safety endpoint (which combined major
promising data from the PLAATO device (Percutaneous bleeding, serious pericardial effusion, and device
Left Atrial Appendage Transcatheter Occlusion; Appriva embolisation) was significantly greater in the device
Medical, CA, USA), this device was withdrawn by the group than in the control group (relative risk 1-69,
manufacturer in 2006.4 95% Bayesian credible interval 101-319), mostly due
Another device that has been developed for LAA to 22 (4-8%) of these patients undergoing either
occlusion is the WATCHMAN system (Aritech Inc, MN, percutaneous or surgical drainage of serious pericardial
USA), which is a self-expanding nitinol frame structure. effusions, none of which was fatal.
Early data on this system were reported in 2007,5 and this Several points arise in light of this study, some of
device is the focus of the recently published PROTECT-AF which might affect future work on atrial fibrillation. Not
(WATCHMAN Left Atrial Appendage System for Embolie all thrombi originate from the LAA, with up to a quarter
Protection in Patients with Atrial Fibrillation) study.6 In of strokes in patients with atrial fibrillation caused
this multicentre, non-inferiority trial, the efficacy and by cerebrovascular disease, complex atheromatous
safety of implantation of the percutaneous LAA closure plaques involving the aorta and carotid arteries, and
www.thelancet.com/neurology Vol 8 October 2009 877