Delayed Choroidal Perfusion in Giant Cell Arteritis

Transcription

1 Tournai of Clinical Neuro-ophthalmology 11(4): , Raven Press, Ltd., New York Delayed Choroidal Perfusion in Giant Cell Arteritis H. G. Mack, B. Me SC., M.B.B.S., J. O'Day, F.R.A.C.P., F.R.A.C.O., and J. N. Currie, F.R.A.C.P. The fundus fluorescein angiograms of 1 patients with visual disturbance due to biopsy-proven giant cell arteritis (11 with anterior ischemic optic neuropathy (AION); 2 with visual obscurations only) were compared with the fluorescein angiograms from patients with acute nonarteritic AION and 2 age-matched normal eyes. In all 1 patients with giant cell arteritis, the fluorescein angiograms showed a significant delay of choroidal filling time (mean 69 seconds) in comparison with either normal subjects (mean 5.8 seconds) or patients with nonarteritic AlaN (mean 5.5 seconds). In patients presenting with acute AlaN, the finding of delayed choroidal filling on fluorescein angiography should raise the index of suspicion of giant cell arteritis and lead to prompt investigation and treatment. Key Words: Temporal arteritis-giant cell arteritis Anterior ischemic optic neuropathy-choroidal blood flow-delayed choroidal filling-fluorescein angiography. Anterior ischemic optic neuropathy (AION) is a common cause of sudden visual loss in the elderly (1-4). Although the majority of cases of AION are nonarteritic, a proportion of cases occur as a complication of giant cell arteritis (GCA) (5). Rapid recognition of arteritic AION and institution of high dosage corticosteroid therapy is a vital part of clinical management, if further visual loss is to be limited or prevented in patients with GCA (6). Unfortunately, in some cases the early differentiation of arteritic from nonarteritic AION may be difficult for reasons that include the absence of prodromal systemic symptoms (7), normal erythrocyte sedimentation rate (ESR) (8,9), delay in obtaining or histologically examining a temporal artery biopsy specimen, and false negative biopsy results due to 'skip lesions' (10). Hayreh has noted that in patients with arteritic AION, there may be delayed choroidal filling on fundus fluorescein angiography (11). To investigate the role of fundus fluorescein angiography in the diagnosis of arteritic AION, we compared the fundus fluorescein angiograms (FFAs) from 1 patients with visual disturbance due to biopsyproven GCA (11 with acute AION; 2 with only transient visual obscurations) with the FFAs from patients with acute nonarteritic AION and 2 aged-matched controls. METHODS From the Neuro-Ophthalmology Clinic, St. Vincent's Hospital, Melbourne, Australia. Address correspondence and reprint reque~ts to Dr. He~ther G. Mack at Howard Florey Institute of Expenmental MedICIne and Physiology, University of Melbourne, Parkville 052, Australia. Retrospective review was made of the clinical records and FFAs from three subject groups: (a) 1 consecutive patients presenting to our neuroophthalmic service with visual disturbance due to biopsy-proven GCA (11 patients with acute AION 221

2 222 H. G. MACK ET AL. and 2 patients with isolated transient visual obscurations); (b) consecutive patients with acute nonarteritic AlaN; (c) 2 normal controls. All 1 patients with biopsy-proven GCA [11 women; mean age 77 ± 6 years standard deviation (SO); age range 66-86] had an elevated Westergren ESR (>25 mm/hr) and prodromal symptoms consistent with GCA (5). The clinical features of these patients are summarized in Table 1. Five patients had bilateral AlaN, 6 had unilateral AlaN, and 2 had only visual obscurations without AlaN. Cases 1 and 2 with GCA are illustrated in Figures 1 and 2. Thirty-three patients had nonarteritic AlaN (15 women; mean age 65 ± 10 years S0, age range years). In 0 patients, the AlaN was unilateral, and nonarteritic AlaN was diagnosed using the clinical and laboratory criteria of normal ESR, the absence of any systemic symptoms of GCA, and the absence of progression of visual loss in either eye. Temporal artery biopsy was performed in only two of these cases and was negative. Three patients had bilateral nonarteritic AlaN on presentation, diagnosed by a normal ESR, the absence of systemic symptoms of GCA, and a negative temporal artery biopsy in each case. The 2 control subjects (12 women) had mean age 60 ± 15 years S0, age range years. In the control subjects FFAs had been performed for investigation of possible retinal disease and had been found to be normal. Fluorescein Angiography FFAs were performed according to routine current techniques (12) and included multiple early exposures to determine the choroidal and retinal perfusion times. For inclusion in ~he s~dy, patients with AlaN or isolated transient VlSUal obscurations required an initial FFA to have been performed within 2 weeks of the onset of visual symptoms. All FFAs were assessed independently by the three authors, without knowledge of patient diagnosis. Choroidal filling time was defined as the time required for filling of all choroidal lobules after fluorescein was first detected in the eye. In most cases consensus for this was easily reached. In the few cases where any difficulty was experienced in determining the exact end point of choroidal filling, the maximum time determined by the three examiners was taken. Follow-up FFAs after corticosteroid treatment were available in 4 of the 1 patients with GCA. These were assessed in the same manner. RESULTS In the patients with nonarteritic AlaN, the mean choroidal filling time was 5.5 ±.1 seconds S0, (range seconds); this did not differ significantly from the choroidal filling time in the 2 TABLE 1. Clinical features of patients with giant cell arteritis Visual field loss Age Visual (complete or ESA Case (years) Sex Involved eye acuity sectorial) (mmlhr) 1 71 F L NPL Complete F L L HM L Complete R days later R CF R Complete F L 20/120 Complete M L CF Complete F R CF Complete F R 20/0 Sectorial M R R NPL R Complete 8 75 F L 4 days later L 20/120 L Sectorial 90 L L 20/20 L Sectorial 50 R 1 day later R 20/15 R Sectorial 9 78 F L LHM L Complete 85 R 4 days later R 20/40 A Sectorial F R HM Complete F L L PL R Complete 117 R 6 days later R CF L Complete F Rand L R 20/0 8 Obscurations only 105 L 20/ F A 20/40 8 Obscurations only 47 NPL, no perception of light; PL, perception of light only; HM, detection of hand motions at a distance of 0 cm; CF, able to count fingers at a distance of 0 em. 8 Visual acuity is reduced because of coincidental nuclear sclerotic lens changes and age-related macular degeneration.

3 FIG. 1. A 71-year-old white woman days after losing acuity in the left eye to no perception of light. A: Red-free photograph showing a pale, slightly swollen left optic disc. B: Fluorescein angiogram 1 second after fluorescein was first detected in the eye, showing marked delay in choroidal filling including optic disc vessels. C: Fluorescein angiogram 14 seconds after fluorescein was first detected in the eye, showing continuing patchy choroidal filling nasal to the optic disc. JClin NeurIHJphthalmol, Vol. 11, No.4, 1991

4 224 H. G. MACK ET AL. A FIG. 2. A 7-year-old white woman who presented to hospital having lost vision in the right eye. She had noted loss of vision in the left eye days previously. Right visual acuity was finger counting at 0 cm and left visual acuity was hand motions at 0 cm. A: Red-free photograph showing edema extending from the disc margin to the macula in the distribution of the left cilioretinal artery. B: Fluorescein angiogram prior to treatment. 14 seconds after first appearance of fluorescein in the eye, showing profound delay in choroidal filling. (Figure continues.) normal subjects (5.8 :!: 1.9 seconds SO; range seconds; p > 0.990, analysis of variance with Scheffe test post hoc comparison). The choroidal filling times for the 1 patients with GCA are summarized in Tables 2 and. Allll patients with arteritic AION showed significantly delayed choroidal perfusion in comparison with both the nonarteritic AION and the control subjects (mean 69 seconds :!: 8 seconds SO, range seconds; p < analysis of variance). Patient age did not contribute to these differences when analvzed as a covariate. In several cases,,- -."- "rl-;:,i >,Ilin:; of choroidal lobules within an eye, and one case had persistent nonfilling of several choroidal lobules, suggesting areas of choroidal infarction. In comparison, central retinal artery filling times did not differ significantly between the GCA patients (mean 17.4 ± 7.8 seconds SO), and nonarteritic AION and control subjects (14. :!:.7 seconds SO; 14.7 :!:.5 seconds SO, p > 0.960, analysis of variance). Both patients with GCA and transient visual obscurations also had delayed choroidal filling times (case 12: 27 seconds; case 1: 2 seconds). In the 4 patients with GCA in whom follow-up FFAs were available, the choroidal filling time had Jell" Nruro-ophthalmol. \ '01. J 1..\,- ~,"",

5 CHOROIDAL PERFUSION IN GIANT CELL ARTERITIS 225 C FIG. 2. (Continued) C: Fluorescein angiogram prior to treatment, 120 seconds after fluorescein was detected in the eye, showing persisting areas of patchy choroidal filling. D: Fluorescein angiogram, 4 years later, demonstrating deep cupping of the optic disc and persistent areas of chorioretinal atrophy, corresponding to areas of poor filling seen in Fig. 2C. Choroidal filling time in all perfused areas was normal. returned to normal (mean 6.2 ± 1.9 seconds, range 4-9 seconds SO), with the earliest time recorded for this being 17 days from the onset of visual symptoms. Figure illustrates choroidal filling time for all subjects. DISCUSSION Anterior ischemic optic neuropathy is generally thought to be caused by impairment of the blood supply to the optic nerve head, which is derived from the posterior ciliary arteries (2). Most nonarteritic cases of AION probably represent local microvascular disease in these posterior ciliary artery branches. However, in cases of AION associated with GCA, there is usually more diffuse arteritic involvement of the posterior ciliary arteries and impairment of the choroidal circulation might also be anticipated. In the normal fluorescein angiogram, fluorescein usually appears in the choroidal circulation approximately 1 second before commencement of filling of the central retinal artery, although this order may sometimes be reversed. Fluorescein appears simultaneously throughout the choroid, in a lobular distribution. Filling is initially patchy because of varying lengths of the short ciliary arteries. How- JClin Neuro-ophthalmol, Vol. 11. No

6 226 H. G. MACK ET AL. 120 U 100 w ~ Ul ~ w... ~ 80 c.:l Z...J...J G: 60...J < ~ a::: 0 I u 20 I I j 0 CONTROLS NON-ARTERITIC GCA GCA AION PRE- TREATMENT POST- TREATMENT FIG.. ever, by the retinal arteriovenous phase, the choroid is usually uniformly filled and exhibits a ground glass appearance due to leakage of fluorescein from the permeable choroidal vessels (1,14). In the current study, choroidal filling time was not significantly different in normal subjects and in patients with nonarteritic AION. However, all patients with arteritic AION showed a marked delay in choroidal filling time, as did the 2 patients with transient visual obscurations and GCA. This delay in choroidal filling was only seen in the acute phase of arteritic AION. Follow-up angiograms performed weeks to months after the onset of AION demonstrated a return to normal choroidal filling times. Case TABLE 2. Choroidal filling time of patients with giant cell arteritis prior to treatment Time of fluorescein angiograms after visual loss (days) o o Choroidal filling time (seconds) Hayreh has recently described massive choroidal nonperfusion in patients with arteritic AION and has suggested that such choroidal abnormalities are extremely rare in nonarteritic AlON (11). Our study provides quantitative confirmation of these observations and suggests that the finding of delayed choroidal perfusion on FFA is of value in the early differentiation of arteritic from nonarteritic AION. In this study, all patients with nonarteritic AION had normal choroidal perfusion. Our subject numbers do not allow us to conclude that in patients presenting with acute AION, GCA can be definitely excluded by the demonstration of normal FFA choroidal perfusion. In addition, the delayed choroidal filling that we have described is not specific to giant cell arteritis and has also been reported in other conditions, including severe hypertension, toxemia of pregnancy, renal failure, disseminated intravascular coagulation, thrombotic thrombocytopenic purpura, and internal carotid artery occlusion (15,16). TABLE. Choroidal filling time of giant cell arteritis patients after treatment Time of fluorescein Choroidal angiogram after filling time Case visual loss (seconds) 1 29 months 6 2 R years 4 L 4 years 9 5 months days 7 J Clin Neuro-ophthalmol. Vol 11. '

7 CHOROIDAL PERFUSION IN GIANT CELL ARTERITIS 227 However, in the common clinical context of a patient presenting with acute AlON, the presence of choroidal hypoperfusion is highly suggestive of GCA. We suggest that fluorescein angiography is a valuable tool in the early investigation of acute AION and that the finding of delayed choroidal perfusion should lead to further urgent investigation and treatment. REFERENCES 1. Boghen DR, Glaser JS. Ischaemic optic neuropathy. The clinical profile and natural history. Brain 1975;98: Hayreh 55. Anterior ischaemic optic neuropathy 1. Terminology and pathogenesis. Br I Ophthalmol 1974;58: Hayreh 55. Anterior ischaemic optic neuropathy 2. Fundus on ophthalmolscopy and fluorescein angiography. Br I Ophthalmol 1974;58: Hayreh 55. Anterior ischemic optic neuropathy. Berlin: Springer-Verlag, 1975: Cullen JF, Coleiro JA. Ophthalmic complications of giant cell arteritis. Sun' Ophthalmol 1976;20:247--{'O. 6. Beri M, Klugman MR. Kohler JA, Hayreh 55. Anterior ischernie optic neuropathy. 7. Incidence of bilaterality and various influencing factors. Ophthalmology 1987;94: Cullen JF. Occult temporal arteritis. A common cause of blindness in old age. Br I Ophthalmol 1967;51: Biller J, Asconape L Weinblatt ME, Toole JF. Temporal arteritis associated with normal sedimentation rate. lama 1982;247: Eshaghian I. Controversies regarding giant cell (temporal, cranial) arteritis. Doc Ophthalmol 1979;47: Klein RG, Campbell Rj. Hunder GG, Carney JA. Skip lesions in temporal arteritis. Mayo Clin Proc 1976;51: Hayreh 55. Anterior ischaemic optic neuropathy. Differentiation of arteritic from non-arteritic type and its management. Eye 1990;4: Schatz H. Fluorescein angiography: basic principles and interpretation. Ryan 5j. ed. In: Retina. Vol 2. St. Louis: CV Mosby, 1989: Gass JDM. Stereoscopic atlas of macular diseases. rd ed. Vol 1. St. Louis: CV Mosby, 1987: Torczynski E. Choroid and suprachoroid. In: Duane TO, Jaeger EA, eds. Biomedical foundations of ophthalmology, Rev. ed. Vol 1. Philadelphia: Harper & Row, 1987: Gaudric A, Coscas G, Bird AC. Choroidal ischaemia. Am I Ophthalmol 1982;94: Kinyoun JL, Kalina RE. Visual loss from choroidal ischemia. Am I OphthalmoI1986;101: 'Clin Neuro-ophthalmol, Vol. 11, No.4, 1991