Journal of Neuro-Ophthalmology, June 1995, Volume 15, Issue 2

A Controlled Trial of Regional Sympatholysis in the Treatment of Photo-oculodynia Syndrome

INTRODUCTION: The mechanism(s) underlying the eye pain syndrome characterized by photo-oculodynia and decreased tears (herein referred to as PODS) is unknown. Postulating a sympathetically maintained pain mechanism, cervical sympathetic ganglion blocks (CSB) were performed in an open-label trial in two patients as a pilot test of our hypothesis. Because these patients experienced clinically dramatic reductions in signs and symptoms, a double-masked controlled trial was initiated. METHODS: With Institutional Review Board approval and written informed consent, four other subjects with PODS who had experienced temporary reduction (> 50%) of symptoms with a single lidocaine CSB were enrolled. A randomized, double-masked series of three CSBs (saline, bupivacaine, lidocaine) was scheduled for each subject 2-7 days apart. RESULTS: CSB with lidocaine and bupivicaine, but not saline, reduced spontaneous pain and light sensitivity and increased production of tears. Symptom reduction lasted for hours to days, extending beyond clinical signs of local anesthetic-induced sympatholysis. It was also observed that sympatholysis extinguished associated blepharospasm in those subjects with this dystonic condition (one subject in the open-label pilot trial and two of three in the controlled trial). CONCLUSIONS: Results of these preliminary data suggest that the sympathetic nervous system is involved in mediating symptoms associated with PODS in certain individuals.

Journal of Neuro- Oplitlmlmology 15( 2): 90- 94, 1995. © 1995 Raven Press, Ltd., New York A Controlled Trial of Regional Sympatholysis in the Treatment of Photo- oculodynia Syndrome Perry G. Fine, M. D., and Kathleen B. Digre, M. D. Abstract: Introduction: The mechanism( s) underlying the eye pain syndrome characterized by photo- oculodynia and decreased tears ( herein referred to as PODS) is un­known. Postulating a sympathetically maintained pain mechanism, cervical sympathetic ganglion blocks ( CSB) were performed in an open- label trial in two patients as a pilot test of our hypothesis. Because these patients experienced clinically dramatic reductions in signs and symptoms, a double- masked controlled trial was initi­ated. Methods: With Institutional Review Board approval and written informed consent, four other subjects with PODS who had experienced temporary reduction (> 50%) of symptoms with a single lidocaine CSB were enrolled. A randomized, double- masked series of three CSBs ( saline, bupivacaine, lidocaine) was scheduled for each subject 2- 7 days apart. Results: CSB with lidocaine and bupivicaine, but not saline, reduced spontaneous pain and light sensitivity and increased production of tears. Symptom reduction lasted for hours to days, extending beyond clinical signs of local anesthetic- induced sympatholysis. It was also observed that sympatholysis extinguished associated blepharospasm in those subjects with this dystonic con­dition ( one subject in the open- label pilot trial and two of three in the controlled trial). Conclusions: Results of these preliminary data sug­gest that the sympathetic nervous system is involved in mediating symptoms associated with PODS in certain individuals. Key Words: Eye pain- Sympatholysis- Photopho­bia- Oculodynia- Blepharospasm. From the Departments of Anesthesiology ( P. G. F.) and Neu­rology ( K. B. D.), University of Utah Health Sciences Center, Salt Lake City, Utah, U. S. A. Address correspondence and reprint requests to Dr. P. G. Fine, Associate Professor, Department of Anesthesiology, Uni­versity of Utah Health Sciences Center, 50 North Medical Drive, Salt Lake City, UT 84132, U. S. A. Photophobia is a general and conventionally used term to denote aversive responses or abnor­mal intolerance to light. Several pathophysiologi­cal states affecting eye structures cause light-induced sensitivity. Iritis and iridocyclitis, both in­flammatory conditions that affect the iris, produce pain when the eye is exposed to light a n d the pupil constricts. This type of pain is ameliorated with pupillary dilation. Chemical or mechanical irrita­tion of the cornea can cause a pain syndrome that may be made worse by light and that is relieved with topical anesthesia. Another type of light sen­sitivity, " dazzling," is due to abnormality of light scatter and lack of adaptation ( 1). Photophobia ac­companies migraine headaches for unknown rea­sons. Migraineurs have a decrease in light pain threshold, and even normal amounts of light can be painful. Little progress has been made to eluci­date further this or other mechanisms underlying light- activated eye pain. In this article, we address a category of idio­pathic chronic eye pain triggered by even minor ocular trauma, when there is no evident ongoing tissue damage or inflammation. This symptom is frequently accompanied by exquisite light ( visual) hypersensitivity, dry eyes, ocular foreign- body sensation, and blepharospasm. No known pathol­ogy exists in the iris or the cornea to account for the pain or other symptoms experienced by these patients. Although the term photophobia is also commonly used to denote this particular group of patients' conditions, we suggest that this nomen­clature is not clinically useful because it serves only to describe a generalized behavioral response to a normally nonnoxious stimulus. Therefore, we dif­ferentiate this group of patients exhibiting the con­stellation of findings listed b y using the t e rm photo-oculodynia syndrome ( PODS). A paucity of epidemi­ological ( prevalence) data exist from which to draw to further describe this group of patients. 90 EFFECT OF SYMPATHOEYSIS ON PHOTO- OCUEODYNIA 91 Hypothesis Reflex sympathetic dystrophy ( RSD), causalgia, and sympathetically maintained pain ( SMP) are names currently given to pain syndromes that seem to be driven by sympathetic activity. The pathophysiology of this subset of neuropathic pain- producing disorders is poorly understood. Most current evidence suggests that there is sen­sitization of central ( spinal) receiving neurons, up-regulation of a- adrenoreceptors at peripheral pri­mary nociceptive afferents, or both, which lead to the sensory abnormalities experienced by affected patients ( 2- 4). Remarkable similarities exist in patients exhibit­ing these sympathetically mediated pain syn­dromes and those with PODS. These are listed in Table 1. Based on this observation, we postulate that if PODS is a sympathetically maintained pathophysiological state, signs and symptoms should abate, at least temporarily, with regional sympatholysis. To test this hypothesis, a two- stage clinical study was initiated. METHODS All subjects involved in every phase of this study voluntarily enrolled after giving written and verbal consent according to the guidelines and with the approval of the University of Utah Insti­tutional Review Board for Human Studies. All pro­cedures were performed in the Pain Management Clinic, University of Utah Health Sciences Center, by the same anesthesiologist ( P. G. F.). A consistent technique of requiring nothing by mouth ( NPO) status for at least 6 h, procuring intravenous ac­cess, continuous monitoring of vital signs ( auto­mated blood pressure ( BP), electrocardiogram ( ECG), pulse oximetry), and performing cervical sympathetic ganglion block ( CSB), was employed. The anterior C- 6 paratracheal approach with a 22- G " B" bevel needle under aseptic conditions TABLE 1. Similarities between PODS and other sympathetically maintained pains 1. Onset triggered by even minor trauma or undeterminable cause. 2. " Burning" quality to spontaneous pain. 3. Symptoms out of proportion to any evident ongoing tissue damage or inflammation ( typical of other neuropathic pains). 4. Pain to a usually nonnoxious sensory stimulus. 5. Dystonia. 6. Emotionality a common hallmark; symptoms often attributed to primary psychological disturbance rather than physically based disorder. PODS, photo- oculodynia syndrome. was used. Digital pressure was applied caudal to the puncture site during injection to promote ce­phalic spread of injectate. Pilot Subjects This stage involved an unmasked local anes­thetic ( 1% lidocaine, 10 ml) unilateral sympathetic block of the cervical sympathetic chain in two sub­jects experiencing PODS ( Table 2). Based on the positive response to sympatholysis in these two individuals, manifest by elimination of spontane­ous pain, light- induced pain, and foreign- body sensation, a placebo- controlled double- masked trial in four additional subjects was initiated. Control Study Subjects experiencing PODS were enrolled and scheduled to undergo screening with an un­masked CSB with 10 ml 1% lidocaine. Those expe­riencing at least 50% reduction in pain would pro­ceed to receive three additional blocks in random order under double- masked conditions, with sub­jects thereby serving as their own controls. Each block would involve 10 ml preservative- free solu­tions consisting of either 1% lidocaine, 0.5% bu-pivacaine, or placebo ( isotonic saline), scheduled 3- 7 days apart. Measures of pain and pain- relief scores ( verbal and visual analog), subjective reports, and objec­tive observations of blepharospasm, presence of Horner's signs, and vital signs would be recorded before the block and at 5, 10, 15, 30, 45, 60, and 90 min after the block. RESULTS All four subjects enrolled in the control study responded to the screening CSB and completed all three subsequent double- masked blocks ( Table 2). Subjective reports of significant reductions in most troublesome symptoms resulted with local anes­thetic injection but not with saline placebo ( Table 3). All local anesthetic blocks were attended by Horner's signs, objectively substantiating sym­patholysis. The most consistent finding was dramatic de­crease in light- induced pain and sensitivity with onset of sympathetic blockade ( Fig. 1). There were no morbid events, and no clinically significant al­terations in vital signs occurred. One subject ( S. T.) underwent pre- and post-block intraocular pressure ( IOP) measurements ( Goldman Tonometer). A 33% reduction from / Neuro- Ophthalmol, Vol. 15, No. 2, 1995 92 P. G. FINE AND K. B. DIGRE TABLE 2. Photo- oculodynia patient profiles Subject Unmasked J. M. F. Q. Age Gender initial trial 65 F 46 M Symptoms Pod, b, d, p, fbs Pod, ha, d, P Double- masked placebo- controlled trial S. M. ST. K. O. K. W. 39 F 57 F 66 M 58 F Pod, d, p, fbs Pod, ha, d, P, b Pod, ha, d, P, b Pod, ha, d, p, b, fbs Duration 8- 10 years 8 months 4 years 4 years 7 years 9 years Trigger Corneal transplant ( Fuch's dystrophy) Corneal transplant ( keratoconus) Chemical keratitis Unknown Unknown Head trauma Pod, photo- oculodynia; d, dry eyes; p, spontaneous burning pain; fbs, foreign body sensation; b, blepharospasm; ha, headache. baseline normal values of IOP occurred after local anesthetic but not after saline injection. The two pilot patients and two of the control study patients decided to continue with two to six additional lidocaine blocks after completing the formal study. One subject ( F. Q.) experienced com­plete resolution of all symptoms, and the other three have experienced > 50% improvement in signs and symptoms at 6- 12 months' follow- up. One subject ( K. O.) did not elect to proceed with any further therapy after the control study ( 3 months' follow- up) because he felt " good enough." The remaining subject, K. W., who had only two local anesthetic blocks during the control trial, at 1 year follow- up reports only the continu­ing problem of dry eyes. DISCUSSION Although the small cohort of subjects under in­vestigation obviates formal statistical analysis, the clinical results demonstrated using this rigorous study design strongly suggest that PODS is in some way driven by sympathetic nervous system activity. This is not surprising because the eye is TABLE 3. Photo- oculodynia patient results: double- masked, placebo- controlled trial of cervical sympathetic ganglion block Result Spontaneous pain Light- triggered pain Blepharospasm Dry eyes Foreign- body sensation S. M. 4a 1 NA 44 4 4 s S. T. 4 44 4 4 4 NA jbject K. O. 4 4 4 4 4 4 NA K. W. 4 4 4 4 4 4 4 4 4 , most remarkable changes perceived by the subjects; NA, not applicable to this subject. " Arrows depict direction and magnitude of change after local an­esthetic block compared to placebo block. richly innervated by sympathetics, and most eye structures are dependent on balanced autonomic activity for normal function. The sympathetics arise in the hypothalamus and travel down the spinal cord to C8- T1 where they leave the cord and synapse in three to four cervical sympathetic ganglia. Fibers from the most cephalic ganglion, the superior cervical ganglion, form the internal carotid plexus. These sympathetic effer-ents enter the orbit at the superior orbital fissure and join the ophthalmic and nasociliary branches of the trigeminal nerve. The sympathetic fibers in the short ciliary nerves supply blood vessels in the eye, whereas the long ciliary nerves supply the dilator of the pupil. placebo local anesthetic time after block FIG. 1. Changes in ocular pain induced by ambient room lighting after cervical sympathetic ganglion block. Values reported are means of the four subjects tested, comparing injection of saline and local anes­thetic. / Neuro- Ophthalmol, Vol. 15, No. 2, 1995 EFFECT OF SYMPATHOEYSIS ON PHOTO- OCUEODYNIA 93 Pharmacologically, sympathetic effector sites consist of several subtypes of receptors. Postsyn­aptic a- 1 receptors are activated by norepineph­rine, whereas presynaptic ct- 2 receptor activation inhibits norepinephrine release from sympathetic efferent terminals. It is well documented that ad­renergic neurotransmitter release can mediate no­ciception, presumably via a- 1 receptor activation of primary nociceptive afferents ( 5,6). p receptors have not been implicated in nociception. Historically, there have been repeated efforts to understand the neural basis of photophobia and eye pain. The sympathetic nervous system or dys-regulation of sympathetic receptors have only in-ferentially been tied to eye pain syndromes. In 1896, Kraus ( 7) performed a seminal experiment on a patient with photophobia whereby he extirpated the gasserian ganglion and sectioned the sensory root of the trigeminal nerve. He reported that symptoms were abolished on the operative side. This led to the belief that photophobia is primarily due to an abnormality in the pain pathways cours­ing with the trigeminal nerve. Later, Lebensohn ( 8) concluded that photopho­bia must be associated with the trigeminal nerve and its interaction with the pupillary light reflex. This investigator subcategorized painful responses to light as " true photophobia" in contradistinction to non- pain- producing sensations such as " daz­zling." He went on to state that for photophobia to occur, the trigeminal nerve must be intact and " an­tidromic vasodilatation" takes place, referred to as a " trigeminal pupillary reflex." He believed that vasodilatation played a fundamental role in the painful response. For photophobia to occur, an ab­normality had to take place in the anterior seg­ment. He suggested that vasodilatation occurred reflexively, causing the subject to perceive an in­creased brightness to the light with consequent pain. Lowenfeld ( 1) has shown that this antidromic vasodilatation actually results from neurotransmit­ter release at terminal effector sites within the eye ( e. g., substance P; prostaglandins), leading to con­gestion of the iris and pupillary miosis. Other possibilities must be taken into account to explain the fact that all PODS patients tested in our study experienced relief with CSB. A placebo re­sponse must always be considered, and this is es­pecially the case when a dramatic intervention such as an injection therapy is performed ( 9). Ver-dugo and Ochoa have repeatedly and appropri­ately emphasized the importance of placebo-controlled methodology in drawing mechanistic conclusions ( 10,11). We believe that a placebo-generated response is an unlikely possibility be­cause the results from placebo ( saline) and active drug ( local anesthetic) injections were so different ( Fig. 1; Table 3). An alternative but related explanation is that these patients suffer from a primary psychiatric disorder that leads to somatization involving this particular area of the body. There are strong argu­ments in the literature linking such attributions to patients with RSD- like syndromes ( 12). In spite of the fact that these patients often respond strongly to suggestion, this primary " causality" argument is countered by other data concluding that the emotionality and affective disorders so frequently observed in this patient population represent co­morbidity ( 13). It has been suggested that this dilemma is best resolved in individual cases by employing a rigor­ous placebo- controlled testing paradigm ( 14). We believe we have fulfilled this criterion for distin­guishing pathophysiological causation in this study. We acknowledge that specific psychological testing may be useful in further defining comor­bidity and directing concurrent psychological ther­apies. Another important consideration is the possibil­ity that systemic absorption of local anesthetic alone caused resolution of symptoms. Intravenous local anesthetic administration has been shown to abate certain neuropathic pains ( 15). To test this possibility, one control study subject ( S. M.) was given the same dose of intravenous lidocaine ( 100 mg) as was injected during the block protocol. This was administered as one of six sequential 15- min intravenous infusions ( five saline, one lidocaine) in a double- masked random order. This followed a previously described methodology specifically de­signed to bring out placebo effects ( 9). Resolution of spontaneous and light- evoked eye pain coin­cided exactly with lidocaine infusion. However, this patient's other symptoms, most notably dry eyes, did not improve as with CSB. This overlap­ping but separation of systemic versus regional sympatholytic effects requires further investiga­tion and confirmation. CONCLUSION A link between the sympathetic nervous system and PODS, an idiopathic pain- producing condi­tion, seems to have been confirmed, but there are considerable barriers to studying patients with PODS to exact a pathophysiological mechanism. Other than serious ethical concerns, which limit human experimentation, the heterogeneous na- / Neuro- Ophtlmimol, Vol. 15, No. 2, 1995 94 P. G. FINE AND K. B. DIGRE ture of these patients in regard to their functional and psychological states makes fixed variable ex­periments and conclusions extremely difficult. This is especially true with an end point as subjec­tive as is pain. Other forms of clinical studies can help to rein­force or discount our conclusion and further define subsets of patients. Previously described method­ologies such as the intravenous phentolamine test and microneurographic tests could be useful ( 16,17). If the sympathetic hypothesis is sup­ported, then better ( i. e., less costly, less invasive, more efficacious) therapeutic strategies must be clinically evaluated. These might include orally available " membrane stabilizing" drugs like mex-iletine or topical ophthalmic solutions to prevent release of norepinephrine ( e. g., clonidine) or block the effects of norepinephrine ( e. g., phentolamine). To study these latter possibilities, an animal model is in development. This is crucial to test behavioral and physiological responses to the spe­cific noxious sensory stimulus ( light) in addition to toxic effects on eye tissues. REFERENCES 1. Lowenfeld I. The " dazzling" syndrome. In: Lowenfeld I, ed. The pupil. Detroit: Wayne State University Press, 1993: 1200- 2. 2. Roberts WJ. A hypothesis on the physiological basis for causalgia and related pains. Pain 1986; 24: 297- 311. 3. Campbell JN, Meyer RA, Raja SN. Is nociceptor activation by alpha- 1 adrenoreceptors the culprit in sympathetically maintained pain? APS } 1992; 1: 3- 11. 4. Fine PG. Pain research and basic science: pain and the sym­pathetic nervous system. In: Stanley TH, Ashburn MA, eds. Anesthesiology and pain management. Boston: Kluwer Ac­ademic Publishers, 1994: 65- 75. 5. Devor M. Nerve pathophysiology and mechanisms of pain in causalgia. / Auton New Syst 1983; 7: 371- 84. 6. Hu SJ, Zhu J. Sympathetic facilitation of sustained dis­charges of polymodal nociceptors. Pain 1989; 38: 85- 90. 7. Lebensohn JE. The nature of photophobia. Arch Ophthalmol 1934; 12: 380- 90. 8. Lebensohn JE. Photophobia: mechanisms and implications. Am J Ophthalmol 1951; 34: 1294- 1300. 9. Fine PG, Roberts WJ, Gillette RG, Child TR. Slowly devel­oping placebo responses confound tests of intravenous phentolamine to determine mechanisms underlying idio­pathic chronic low back pain. Pain 1994; 56: 235- 42. 10. Verdugo RJ, Ochoa JL. Use and misuse of conventional electrodiagnosis, quantitative sensory testing, thermogra­phy, and nerve blocks in the evaluation of painful neuro­pathic syndromes. Muscle Nerve 1993; 16: 1056- 62. 11. Verdugo RJ, Ochoa JL. Sympathetically maintained pain. Neurology 1994; 44: 1003- 9. 12. Egle UT, Hoffman SO. Psychosomatic aspects of reflex sympathetic dystrophy. In: Stanton- Hicks M, Janig W, Boas RA, eds. Reflex sympathetic dystrophy. Boston: Kluwer Academic Publishers, 1990: 29- 36. 13. DeGood DE, Cundiff GW, Adams LE, Shutty MS. A psy­chosocial and behavioral comparison of reflex sympathetic dystrophy, low back pain, and headache patients. Pain 1993; 54: 317- 22. 14. Ochoa JL. Essence, investigation, and management of " neuropathic" pain; hopes from acknowledgment of chaos. Muscle Nerve 1993; 16: 997- 1008. 15. Galer BS, Miller KV, Rowbotham MD. Response to intra­venous lidocaine infusion differs based on clinical diagno­sis and site of nervous system injury. Neurology 1993; 43: 1233- 5. 16. Raja SN, Treede R- D, Davis KD, Campbell JN. Systemic alpha- adrenergic blockade with phentolamine: a diagnostic test for sympathetically maintained pain. Anesthesiology 1991; 74: 691- 8. 17. Murakami K, Sobue G, Iwase S, Mitsuma T, Mano T. Skin sympathetic nerve activity in acquired idiopathic general­ized anhidrosis. Neurology 1993; 43: 1137- 40. / Neuw- Ophthahnol, Vol. 25, No. 2, 1995