Letter to the Editor |
Intractable cluster headache and therapeutic stimulation of the hypothalamus: pathophysiological and management insights from a rare experiment
Dubai Police Medical Services, P.O. Box 12005, Dubai, United Arab Emirates
E-mail: docgupta{at}emirates.net.ae; dr_vkgupta{at}yahoo.com
Received September 3, 2004. Revised November 11, 2004. Accepted November 29, 2004.
Sir,
Leone et al. (2004)
present the first patient whose cluster headache (CH) attacks were controlled by bilateral hypothalamic stimulation. In-depth study of this rare occurrence offers an unusual opportunity to discuss certain key aspects surrounding CH management and pathophysiology.
Clinical experience indicates that despite a negative chronotropic action, verapamil is probably the most effective agent in the treatment of CH; surgical procedures for CH should be considered only following total resistance to medical therapy (Mathew, 1990). Surprisingly, Leone et al. (2004) did not attempt CH prophylaxis with verapamil in this patient at any time. This therapeutic strategy would be particularly indicated in patients with bilateral CH attacks associated with hypertensive crises. To label a CH patient as being ‘intractable’ without attempting prophylaxis with verapamil is questionable. The only contraindication to verapamil prophylaxis in this case might be the relative bradycardia; however, bradycardia around 40 beats per minute is not unusual in CH (Attanasio et al., 1990; Mathew, 1990). In this athletic subject with physiological vagotonia, bilateral hypothalamic stimulation itself caused bradycardia sufficient to induce vertigo (Leone et al., 2004), possibly due to diminished cardiac output and brain perfusion. Secondly, there is no evidence of target organ damage (cerebral, cardiac or renal) to support the belief that the hypertensive crises were ‘life-threatening’ in this case, another justification for invasive experimental therapy. These hypertensive crises might represent an unusual variant of central neurogenic hypertension linked to the CH attacks. Remarkably, reduction of calcitonin gene-related peptide (CGRP) can contribute to hypertensive crises (Deng et al., 2004). Extensive palpebral oedema could induce a selective depletion of CGRP that, in turn, predisposed to the hypertensive crises seen in this case. CGRP depletion might also be involved in the milder hypertension commonly seen (Attanasio et al., 1990) in CH attacks.
Activation of pain-modulating pathways by hypothalamic stimulation, as suggested (Leone et al., 2004), does not appear to subserve a key component of the CH-remitting mechanism in this case, because the headache attacks also involved prominent palpebral oedema as well as episodic hypertension. Additionally, neither verapamil nor indomethacin freely crosses the intact blood–brain barrier; such drugs are unlikely to influence brain function either at the level of the cortex or the brainstem, including antinociceptive mechanisms (Gupta, 2003). In addition, lithium does not influence antinociceptive mechanisms. Conversely, methylprednisolone (bolus 125 mg) appears to have a sustained analgesic effect (Romundstad et al., 2004), but bolus steroids showed only transient benefit in this case. These pharmacological limitations challenge the theory of a generalized defect in antinociception in primary headaches, including the observation of reduced pain pressure threshold in CH (Antonaci et al., 1993). Lateralization of primary headache (migraine, CH, chronic paroxysmal hemicrania or hemicrania continua), as well as unpredictable onset or remission of headaches, are additional concerns with the concept of defective antinociception as a persistent or constitutional trait (Gupta, 2003). The significantly wide lag (median 4 days; range 1–46 days) between onset of high frequency hypothalamic stimulation and disappearance of the crises in this case complicates the interpretation. The lag between turning off the stimulators and attack recurrence is even wider (mean 72.8 days; range 2–290 days). A tardy resetting of the hypothalamus and/or other brain function is the only speculative mechanistic possibility, as has been considered by Leone et al. (2004). Nevertheless, which brain function(s) is being affected by such electrode stimulation is far beyond speculation. In most labile clinical circumstances with rather soft subjective end points and high-profile treatments, including coronary artery bypass surgery (Johnson, 1994), the placebo effect cannot easily be excluded.
The precise role of surgical manipulations of the trigeminal nerve (alcohol injections, balloon compressions, thermal or radiofrequency or surgical rhizotomies) remains uncertain, particularly in the face of recurrence of CH after induction of complete trigeminal anaesthesia, as also seen in this case (Leone et al., 2004). If, as is generally agreed, the trigeminal nerve fibres constitute the final common pathway for expression of pain and autonomic symptoms of CH (Mathew, 1990), it is logically challenging to maintain that persistence or recurrence of attacks after surgical manipulations on the trigeminal nerve supports (Matharu and Goadsby, 2002) the concept of central brain neuronal origin of CH, particularly in view of the significant initial benefit seen following the surgical procedures. Furthermore, therapeutic successes with verapamil and indomethacin in preventing and aborting CH, respectively, unambiguously indicate a peripheral origin for this primary headache. Thirdly, the suggestion that excellent response to sumatriptan post-trigeminal nerve section supports a central neuronal origin of CH (Matharu and Goadsby, 2002) seems pharmacologically invalid. The brain penetrability of sumatriptan is poor, suggesting negligible neuronal influence (Hargreaves, 2004). Finally, alcohol, sublingual nitroglycerine or subcutaneous histamine can provoke CH (Mathew, 1990), but no aberration of brain neuronal function clearly linked to any of these pharmacological agents has been implicated in the pathogenesis of CH. A pharmacological overview therefore finds little in direct support for a central or brain neuronal origin of CH. It is, nevertheless, acceptable that following complete section of the trigeminal sensory root, the brain would not perceive either vasodilatation or a peripheral neural inflammatory process through the usual neuro-anatomic pathway (Matharu and Goadsby, 2002). To explain perception of pain and pain-associated phenomena in attacks of CH after effective or near-total interruption of afferent neuronal traffic, we have to conceive of alternative neural routes that develop aberrantly after trigeminal nerve section or surgical procedure. The trigeminal nerve is peripherally connected to the III, IV, VI and VII nerves (Williams et al., 1989). Aberrant conduction to the brainstem may set up during CH attacks through some of these neural connections after functional or structural interruption of the trigeminal nerve.
For hypothalamic stimulation to be scientifically accepted as an evidence-based therapeutic option, as well as to elucidate further the pathophysiology of CH, the challenges offered by the basic sciences must be met squarely.
References
Antonaci F, Sandrini G, Pucci E, Arrigo A, Micieli G, Nappi G. Lateralization of pain pressure threshold in cluster headache. Cephalalgia 1993; 13 (Suppl 13): 33.[CrossRef][Web of Science][Medline]
Attanasio A, Capria A, Quatrana M, Piccolo CG, Bisceglia A, Formisano R, et al. Sinus bradycardia, junctional rhythm and blood pressure increase during repeated cluster headache attacks. Headache 1990; 30: 509–10.[Medline]
Deng PY, Ye F, Cai WJ, Deng HW, Li YJ. Role of calcitonin gene-related peptide in the phenol-induced neurogenic hypertension in rats. Regul Pept 2004; 119: 155–61.[CrossRef][Medline]
Gupta VK. Migraine following haemorrhage in brain stem cavernous angioma: pathophysiological considerations. J Neurol Neurosurg Psychiatry. Published online June 23, 2003. Available at: http://jnnp.bmjjournals.com/cgi/eletters/74/5/680#55
Hargreaves R. New concepts in migraine—old things explained in new ways? Headache 2004; 44: 736–8.
Johnson AG. Surgery as a placebo. Lancet 1994; 344: 1140–2.[CrossRef][Medline]
Leone M, Franzini A, Broggi G, May A, Bussone G. Long-term follow-up of bilateral hypothalamic stimulation for intractable cluster headache. Brain 2004; 127: 2259–64.
Matharu MS, Goadsby PJ. Persistence of attacks of cluster headache after trigeminal nerve root section. Brain 2002; 125: 976–84.
Mathew NT. Advances in cluster headache. Neurol Clin 1990; 8: 867–90.[Medline]
Meyer JS, Dowell R, Mathew N, Hardenberg J. Clinical and hemodynamic effects during treatment of vascular headache with verapamil. Headache 1984; 24: 313–21.[Medline]
Romundstad L, Breivik H, Niemi G, Helle A, Stubhaug A. Methylprednisolone intravenously 1 day after surgery has sustained analgesic and opioid-sparing effects. Acta Anaesthesiol Scand 2004; 48: 1223–31.[Web of Science][Medline]
Williams PL, Warwick R, Dyson M, Bannister LH, editors. Gray's anatomy, 37th edn. Edinburgh: Churchill Livingstone; 1989. p. 1096–107; 1120.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||