Medical Hypotheses (2004) 62, 45–48
http://intl.elsevierhealth.com/journals/mehy
Is benign cough headache caused by intraocular haemodynamic aberration? Vinod Kumar Gupta* Dubai Police Medical Services, P.O. Box 12005, Dubai, United Arab Emirates Received 28 April 2003; accepted 15 October 2003
Summary Rise of intracranial pressure (ICP) is currently presumed to underlie benign cough headache (BCH). Cough normally increases ICP but very few patients develop BCH. Children, young adults and females are rarely affected. Reduction of ICP by lumbar puncture (LP) or indomethacin offers variable therapeutic success. BCH can persist for several months or years but LP lowers ICP for few hours only and has significant morbidity. Choroidal blood volume and intraocular pressure (IOP) are instantaneously responsive to cough. Mechanical deformation of pressure-sensitive ocular structures by sudden experimental IOP elevation generates transient neural traffic in ocular trigeminal nerve fibres. Homeostatic mechanisms normally limit effect of cough-induced intraocular venous congestion. I propose that in a few patients, ocular sympathetic hypofunction significantly alters intraocular pressure–volume relation and predisposes to exaggerated choroidal venous congestion and fluctuation of IOP in response to cough, that, in turn, results in sudden transient cephalalgogenic antidromic trigeminal nerve discharge. Known variations in ocular hydrodynamics, ocular rigidity, and forced expiration rationalize epidemiology of BCH. This hypothesis can be tested by study of pupillary function and facial sweating in patients with BCH. c 2003 Elsevier Ltd. All rights reserved.
Introduction Cough-induced rise in intracranial pressure (ICP) is a well-known physiological phenomenon [1]. Sudden increase of ICP [2–4] or intracranial venous pressure [5] is believed to underlie benign cough headache (BCH). BCH is characterized by bilateral headache of sudden onset precipitated by coughing and lasting less than 1 min; the headache develops in absence of any underlying intracranial disorder [6]. Reported benefit of reduction of ICP by lumbar puncture (LP) or administration of indomethacin [3,7–9] appears to support pathogenetic link between BCH and sudden rise of ICP. Valve-like
*
Tel.: +971-4-3355954; fax: +971-4-3344951. E-mail address:
[email protected] (V.K. Gupta).
blockage at foramen magnum producing craniospinal pressure dissociation [1], heightened sensitivity of ICP-sensitive brain receptors [3], and incompetent or absent jugular venous valves [4] have been proposed. It is not known why LP should terminate BCH [10]. Nevertheless, therapeutic LP is currently recommended to manage BCH [3]. Eight of 14 patients failed to respond to LP [3]. Previously, LP effected dramatic response in 1 of 28 patients [7]. Depending upon the nature of cough, BCH can persist for several months or years [3,9]. LP, however, lowers ICP for a few hours. A single spinal tap cannot be expected to sustain long-term remissions of BCH. On the other hand, morbidity of LP includes serious complications such as post-LP headache and spinal canal infections [11]. Despite striking ICP reducing influence [12], indomethacin offers variable therapeutic benefit
0306-9877/$ - see front matter c 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0306-9877(03)00298-6
46
[3,9,13]. The most common central nervous system side-effect of long-term indomethacin administration is severe frontal headache; also, indomethacin has non-specific analgesic properties distinct from its anti-inflammatory effects [14]. Exertional migraine, benign orgasmic cephalalgia, benign exertional headache, chronic paroxysmal hemicrania, and “ice-pick” headache are indomethacinresponsive [5,9,15]. However, neither headacheremitting nor headache-provoking mechanism of indomethacin is understood. Also, indomethacin can itself cause benign intracranial hypertension [16]. Results of propranolol therapy in BCH patients are equivocal [13] or negative [9,17]. Propranolol does not influence ICP. Methysergide variably affects BCH [9,10]. In direct contrast to indomethacin, methysergide increases cerebral blood flow [18]. Indomethacin, methysergide, and propranolol variably affect cerebral blood flow and ICP. Finally, pneumoencephalography (PEG) (now obsolete) was reported to remit BCH [7]. PEG consistently produced a severe headache [19]. LP and PEG effect opposite changes in cerebrospinal fluid (CSF) pressure. Cough normally increases ICP in all persons. Very few persons would never experience cough in their lifetime. Nevertheless, incidence of BCH is strikingly low [3,13,20]. BCH spares both paediatric and young adult populations [3,20]. Predominance of male subjects [3,21,22] is also unexplained. Clinical states with paroxysmal strenuous coughing such as pertusis, asthma or chronic obstructive pulmonary disease generally do not manifest any association with BCH. What is the mechanistic basis for such “protection” from BCH? No variables that might modulate cough-induced physiological elevation of ICP are known. Critical differences exist between BCH and symptomatic cough-induced headache. Coughinduced rise of ICP can be more prolonged and partially cumulative as manifested by increasing frequency, duration and severity of headache attacks in Chiari malformation (type I) [23,24]. Headache characteristics, however, remain unchanged in BCH over extended periods of time. Also, cough-induced headache is not an early feature of raised intracranial tension but is manifest in patients with the later form of continuous (nonparoxysmal) headache [25]. Current theories of BCH assume that in an otherwise normal subject, cough would elicit a CSF impulse of sufficient magnitude to instantaneously cause headache by traction and pressure on pain-sensitive CNS structures. It is inappropriate to extend the pathophysiology of symptomatic cough-induced headache to BCH.
Gupta
Hypothesis: intraocular pressure– volume relation and benign cough headache Cough is a forcible explosive expiration with maximally positive intrathoracic and intraairway pressures generated by muscle contraction against a closed glottis [26]. Increase in intrathoracic pressure during coughing or Valsalva manuvre impedes right atrial venous return that, in turn, increases central venous pressure. Cranial venous return is impeded during coughing; cranial venous congestion elevates episcleral venous pressure and congests choroidal vasculature thereby increasing rate of secretion of aqueous humour that, in turn, increases IOP [26,27]. Sneezing and laughing also briefly elevate IOP [26]. The choroid possesses the greatest blood supply in the human body, with a circulation volume 10–20 times that of the cerebral cortex [28]. The trigeminal axon reflex (antidromic discharge) is the key feature in the effector limb of headache [29]. Acute experimental IOP elevations discharge impulses in iris, corneo-scleral, and whole nerve ocular trigeminal fibres probably due to mechanical distortion of iris and chamber angle, suggesting production of painful antidromic impulses [26,30]. Significantly, such stimulation of ocular structures appears to be involved in only transient pain sensations in glaucoma patients [30]. I propose that in BCH patients, a regional ocular sympathetic nervous system insufficiency contributes to inability of the eye to handle sudden surges in intraocular volume and IOP that occur due to congestion in the choroidal venous circulation during coughing. Such exaggerated sudden fluctuation of IOP generates a transient antidromic trigeminal nerve discharge through mechanical stimulation of pressure-sensitive intraocular sensory nerve-endings that in turn precipitates BCH. Evidences from vascular headache research support two components of this hypothesis: (i) Regional ocular and cervical sympathetic deficiency is known to prevail in migraine and cluster headache [5]. (ii) Sudden vasodilatation in intraocular circulation has been documented in cluster headache and chronic paroxysmal hemicrania [31,32]. BCH is nothing but a form of vascular headache with a markedly shortened time frame. The critical conceptual difference that must, however, be maintained between BCH and cluster headache/chronic paroxysmal hemicrania is that in BCH the sudden intraocular vascular congestion is of venous rather than arterial origin. Three homeostatic mechanisms probably affect cough-induced surge of venous blood into the eye:
Is benign cough headache caused by intraocular haemodynamic aberration?
(i) autonomic nervous system innervation of choroidal circulation; (ii) ocular rigidity; and (iii) tamponade effect of IOP. Although the issue is controversial, some studies indicate that the choroid is capable of vigorous autoregulation [33]. Normally, choroidal congestion during cough is likely attenuated partially by dissipation of venous impulse in the sympathetically innervated choroidal venous circulation. Secondly, ocular rigidity is determined by several factors including size of globe, distensibility of corneo-scleral envelope, effect of previous distention, age, cushioning effect of uveal circulation, state of hydration, and temperature [26,27]. Whereas ocular rigidity has little influence on IOP level in slow or “steady-state alterations”, it determines magnitude of acute transient changes in IOP as may be caused by rapid injection of fluid into the eye or its displacement by impression tonometry [26,27]. Precisely such rapid but transient alteration in ocular haemodynamics is envisaged in this hypothesis. Relatively rigid eyes that have limited potential to accommodate cough-induced surge of intraocular blood volume would generate more readily a transient cephalalgogenic trigeminal nerve discharge. Finally, IOP itself exerts an important intraocular tamponade function, with increasing IOP decreasing pressure-sensitive ultrafiltration component of aqueous humour formation [34] and ocular blood volume [35]. Cough-induced venous surge in choroidal blood volume and rise of IOP is, therefore, usually self-limited. Against this background, the rare occurrence of BCH [3,20–22] can be rationalized. Greater distensibility of corneo-scleral envelope in early years [26,27] probably underlies the infrequency of BCH in paediatric and young adult populations. Consequently, cough-induced choroidal venous impulses may be dissipated more completely in younger subsets. Relative resistance of females to develop BCH possibly relates to the fact that cough in females is less forceful. Spirometric measurements made during the forced expiratory manuvre are affected, besides other factors, by sex [36]. The magnitude of chest muscle contraction-dependent choroidal venous impulse during cough is likely lower in females than in males. In children, young adults, and females, trigeminal antidromic activation due to cough-induced choroidal congestion may generally remain below the threshold required to precipitate BCH. To test this hypothesis, cranial autonomic function studies should be performed in BCH patients. Pupillary and facial sweating functions can unravel the proposed oculo-cranial sympathetic deficit.
47
Conclusion Since LP was first used as a therapeutic measure in BCH almost 50 years ago, no meaningful progress has been made in comprehension of underlying pathophysiology. We have naturally but simplistically assumed that the physiological cough-induced impulse in CSF pressure causes BCH. In-depth analysis of available evidences does not support the presumed link between BCH and ICP. It should no longer be considered ethical to offer empirical invasive potentially morbid therapy such as LP for BCH. IOP is immediately responsive to cough and Valsalva manuvre, a function of the highly vascular and labile choroidal venous circulation. Cough-induced alteration in intraocular pressure–volume relation is presented as a new pathogenetic hypothesis for BCH. Adaptive distensibility of the corneo-scleral envelope declines with advancing age, thereby setting the stage, in combination with regional ocular sympathetic nervous system hypofunction and choroidal venous congestion, for occurrence of BCH. In the absence of any underlying cranial anomaly, other activities that suddenly increase intrathoracic pressure such as sneezing, laughing, stooping, lifting, straining at stools probably cause short-lasting headache by a similar mechanism.
References [1] Williams B. Cerebrospinal fluid pressure changes in response to coughing. Brain 1976;99:331–46. [2] Williams B. Cough headache due to craniospinal pressure dissociation. Arch Neurol 1980;37:226–30. [3] Raskin NH. The cough headache syndrome: treatment. Neurology 1995;45:1784. [4] Knappertz VA. Cough headache and the competency of jugular venous valves. Neurology 1996;46:1497. [5] Lance JW. Solved and unsolved headache problems. Headache 1991;31:439–45. [6] Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia 1988;8(Suppl. 7):1–96. [7] Symonds C. Cough headache. Brain 1956;79:557–68. [8] Rooke ED. Benign exertional headache. Med Clin North Am 1968;52:801–8. [9] Mathew NT. Indomethacin-responsive headache syndromes. Headache 1981;21:147–50. [10] Bahra A, Goadsby PJ. Cough headache responsive to methysergide. Cephalalgia 1998;18:495–6. [11] Evans RW. Complications of lumbar puncture. Neurol Clin 1998;16:83–105. [12] Harrigan MR, Tuteja S, Neudeck BL. Indomethacin in the management of elevated intracranial pressure: a review. J Neurotrauma 1997;14:637–50. [13] Calandre L, Hernandez-Lain A, Lopez-Valdes E. Benign Valsalva’s maneuver-related headache: an MRI study of six cases. Headache 1996;36:251–3.
48
[14] Gilman AG, Rall TW, Nies AS, Taylor P, editors. Goodman and Gilman’s the pharmacological basis of therapeutics. 8th International ed. New York: McGraw-Hill; 1992. p. 659–61. [15] Sands GH, Newman L, Lipton R. Cough, exertional, and other miscellaneous headaches. Med Clin North Am 1991;75:733–47. [16] Konimi H, Imai M, Kamoshitas NK, Tada H. Indomethacin causing pseudotumor cerebri in Barter’s syndrome. N Engl J Med 1978;298:855. [17] Hazelrigg RL. IV DHE-45 relieves exertional cephalgia. Headache 1986;26:52. [18] Hadjiev D. Impedance methods for investigation of cerebral circulation. Brain Res 1974;14:25–85. [19] Walton JN, editor. Brain’s diseases of the nervous system. 8th ed. Oxford: Oxford University Press; 1981. p. 264. [20] Rasmussen BK, Olesen J. Symptomatic and nonsymptomatic headaches in a general population. Neurology 1992; 42:1225–31. [21] Pascual J, Iglesias F, Oterino A, Vasquez-Barquero A, Berciano J. Cough, exertional, and sexual headaches: an analysis of 72 benign and symptomatic cases. Neurology 1996;46:1520–4. [22] Pascual J, Berciano J. Experience in the diagnosis of headaches that start in elderly people. J Neurol Neurosurg Psychiatry 1994;57:1255–7. [23] Negoro K, Tsuda N, Morimatsu M, Kurokawa Y, Abiko S. A case of cough headache with Chiari malformation (type 1). Rinsho Shinkeigaku 1993;33:327–30. [24] Stovner LJ. Headache associated with the Chiari type 1 malformation. Headache 1993;33:175–81. [25] Bannister R, editor. Brain’s clinical neurology. 6th ed. Oxford: Oxford University Press; 1984. p. 229.
Gupta
[26] Duke-Elder S. System of ophthalmology. The physiology of the eye and of vision, vol. 4. London: Henry Kimpton; 1968. p. 276–7; 280–3. [27] Shields MB. Textbook of glaucoma. 2nd ed. Baltimore: Williams & Wilkins; 1987. p. 45–70. [28] Bill A. Blood circulation and fluid dynamics in the eye. Physiol Rev 1975;55:383–417. [29] Moskowitz MA, Reinhard JF Jr, Romero J, Melamed E, Pettibone DJ. Neurotransmitters and the fifth cranial nerve: is there a relation to the headache phase of migraine? Lancet 1979;ii:883–5. ~ez J, Belmonte C. Sensory nerve responses [30] Zuazo A, Iban elicited by experimental ocular hypertension. Exp Eye Res 1986;43:759–69. [31] Hørven I, Sjaastad O. Cluster headache syndrome and migraine: ophthalmological support for a two-entity theory. Acta Ophthalmol Copenhagen 1977;55:35–51. [32] Sjaastad O, Egge K, Hørven I, Kayed K, Lund-Roland L, Russell D, Slørdahl Conradi I. Chronic paroxysmal hemicrania: mechanical precipitation of attacks. Headache 1979;19:31–6. [33] Kiel JW. The effect of arterial pressure on the ocular pressure–volume relationship in the rabbit. Exp Eye Res 1995;60:267–78. [34] Krupin T. Aqueous humour dynamics. In: Easty DL, Sparrow JM, editors. Oxford textbook of ophthalmology, vol. 1. New York: Oxford University Press; 1999. p. 629–39. [35] Broadfoot K, Gloster J, Greaves DP. Photoelectric method of investigating the amount and oxygenation of blood in the fundus oculi. Br J Ophthalmol 1961;45:161–82. [36] Gold WM, Boushey HA. Pulmonary function testing. In: Murray JF, Nadel JA, editors. Textbook of respiratory medicine. Philadelphia: Saunders; 1988. p. 611–82.
