Use of Retinal Angiography and MRI in the Diagnosis of Giant Cell Arteritis With Early Ophthalmic Manifestations

Background: Giant cell arteritis (GCA) is a vasculitis often revealed by visual signs. Diagnosis is challenging and urgent. Retinal angiography (RA) and MRI allow effective diagnosis. We compared those and proposed an imaging-based approach to diagnose GCA in ophthalmological practice. Methods: We conducted a retrospective study based on the data collected from patients suspected to have GCA on ophthalmological findings. Fluorescein (FA) and indocyanine green (ICG) RAs and MRI were performed and compared with final diagnosis. Results: Among the 41 patients included, 25 were diagnosed with GCA. Sensitivities and specificities of FA and ICG were not different. MRI showed a higher sensitivity and specificity. The approach consisting in performing RA followed by MRI provided a better accuracy. Conclusion: Our study shows that RA can be supplemented by MRI in a specialized center to provide the most accurate diagnosis in GCA revealed by visual signs.

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Use of Retinal Angiography and MRI in the Diagnosis of Giant Cell Arteritis With Early Ophthalmic Manifestations Alexandre Dentel, MD, Gaelle Clavel, MD, Julien Savatovsky, MD, Catherine Vignal, MD, Thomas Senè, MD, Frédérique Charbonneau, MD, Kevin Zuber, MSc, Augustin Lecler, MD, Rabih Hage, MD Background: Giant cell arteritis (GCA) is a vasculitis often revealed by visual signs. Diagnosis is challenging and urgent. Retinal angiography (RA) and MRI allow effective diagnosis. We compared those and proposed an imagingbased approach to diagnose GCA in ophthalmological practice. Methods: We conducted a retrospective study based on the data collected from patients suspected to have GCA on ophthalmological findings. Fluorescein (FA) and indocyanine green (ICG) RAs and MRI were performed and compared with final diagnosis. Results: Among the 41 patients included, 25 were diagnosed with GCA. Sensitivities and specificities of FA and ICG were not different. MRI showed a higher sensitivity and specificity. The approach consisting in performing RA followed by MRI provided a better accuracy. Conclusion: Our study shows that RA can be supplemented by MRI in a specialized center to provide the most accurate diagnosis in GCA revealed by visual signs. Journal of Neuro-Ophthalmology 2022;42:218–225 doi: 10.1097/WNO.0000000000001517 © 2022 by North American Neuro-Ophthalmology Society G iant cell arteritis (GCA) is a vasculitis involving largesized and medium-sized arteries (1). It is the most common systemic vasculitis in the elderly (2,3) with a mean age of 71 years at diagnosis (4). The sex ratio is approximately 7 women for 3 men (5). The cranial form of GCA Department of Ophthalmology (AD, CV, RH), Foundation Adolphe de Rothschild Hospital, Paris, France; Department of Internal Medicine (GC, TS), Foundation Adolphe de Rothschild Hospital, Paris, France; Department of Neuroradiology (JS, FC, AL), Foundation Adolphe de Rothschild Hospital, Paris, France; and Department of Clinical Research (KZ), Foundation Adolphe de Rothschild Hospital, Paris, France. The authors report no conflicts of interest. Address correspondence to Alexandre Dentel, MD, Department of Ophthalmology, Foundation Adolphe de Rothschild Hospital, Paris, France; E-mail: [e-mail]. 218 (C-GCA) is the most common in ophthalmology clinics (6). By affecting arteries supplying the eyes and the cranial nerves, C-GCA can induce a variety of visual manifestations in up to 68% of cases (7–11). In patients with GCA with optic nerve ischemia, visual prognosis is poor, with a high prevalence of irreversible visual loss (12,13). Emergent diagnosis and treatment are therefore needed (14) because bilateralization concerns up to 54% of patients within the few days after the first eye involvement (15–17). Several diagnosis tools are available. Blood testing allows screening for signs of inflammation (increased C-reactive protein [CRP] or erythrocyte sedimentation rate [ESR]) (5). Histopathological evidence with temporal artery biopsy (TAB) remains widely performed as a gold standard test in GCA and can reveal segmental and focal nonnecrotizing giant cell panarteritis (18). However, numerous studies showed lower diagnostic accuracies of TAB in GCA when compared with imaging (19–23). Several studies showed that contrastenhanced vessel-wall MRI (CE-VW MRI) (referred below as “MRI”) can analyze parietal inflammation in cephalic arteries, allowing effective diagnosis in GCA (24–28). Better resolution and diagnostic accuracy were demonstrated using 3 T units (29–33). Besides, a strong superiority of “reformated” 3D MRI over “classic” 2D MRI has been proven, reaching 100% specificity (34). However, MRI, ultrasound, and even TAB need significant resources and trained operators to provide reliable results. They are not routinely available to every ophthalmologist suspecting GCA. Fluorescein (FA) or indocyanine green (ICG) retinal angiographies (RAs) are more widely available examinations in ophthalmological primary care centers or clinics. Most ophthalmologists are trained to identify typical angiographic findings in GCA, that is, choroidal circulatory filling delay related to posterior ciliary network ischemia (35–37). The goal of this study was to assess the use of FA, ICG, and MRI for diagnosing GCA in patients with suspicious Dentel et al: J Neuro-Ophthalmol 2022; 42: 218-225 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution diplopia or visual loss. We will (1) establish the sensibilities and specificities of these 3 examinations and (2) propose an imaging-based approach to diagnose GCA in ophthalmological practice. METHODS Patients We conducted a retrospective single-centered study at the Hôpital Fondation Ophtalmologique Adolphe de Rothschild (Paris, France), based on data collected from the HORTIM cohort. From December 2014 to September 2017, the HORTIM cohort prospectively included patients who (1) were at least 50 years; (2) were clinically suspected of having GCA; and (3) underwent several imaging examinations including FA, ICG, and MRI. Patients were not included in the HORTIM cohort if (1) they were already given a diagnosis of GCA and (2) TAB or final diagnosis preceded imaging. We excluded HORTIM cohort’s patients from this study if they (1) did not initially complained about visual loss or diplopia; (2) did not undergo FA, ICG, MRI, or TAB; and (3) had an ocular or systemic condition that could interfere with RAs lecture (see below). On admission, each patient underwent (1) physical examination by an internal medicine specialist; (2) comprehensive ophthalmic examination including visual acuity measurement, eye motility examination; (3) nonmydriatic fundus photograph; and (4) measurement of ESR and CRP rates. Each patient was classified according to the visual impairment(s) they presented at baseline, deemed “afferent” if there was loss of vision caused by ischemia of the retina or the optic nerve, “efferent” if there was binocular diplopia related to cranial nerve palsy, and “mixed” when both loss of vision and diplopia were reported (Table 1). Clinical and paraclinical findings were reviewed by 2 neuro- ophthalmologists. Two follow-up clinical examinations were performed in the neuro-ophthalmology department (at 2 and 6 weeks after presentation). Retinal Angiography Each RA included at least 8 images between the early and late phases of the angiogram, both dyes collected at identical time intervals, captured on SPECTRALIS Optical Coherence Tomography (OCT) Angiography Module (Heidelberg Engineering). Their analysis was performed by experienced retinologists, blind to the final diagnosis. RA was considered positive if it showed extensive or sectorial choroidal circulation filling delay in either eye at inclusion. A difference of artery filling by 20 seconds between the 2 eyes (or between 2 sectors of the same eye) was considered abnormal (Fig. 1). The presence of other causes of choroidal ischemia, such as uveal effusion syndrome, posterior scleritis/inflammation, ipsilateral highgrade stenosis of the internal carotid artery, severe hypertension, and shock (39), was excluded. MRI MRI was performed on a Philips Ingenia 3 T (Philips Medical Systems), with systematic intravenous injection of gadobutrol. The results were blindly analyzed by 2 neuroradiologists (AL and FC) because it has been described and recommended (24,38). Each MRI was scored on 3 levels: certain (Fig. 2), suspicious, or negative, following previously referenced keys. In this study, we considered MRI as positive if scored as certain and negative if not. Temporal Artery Biopsy and Final Diagnosis The final diagnosis of GCA was assessed by a multidisciplinary team of rheumatologists and internists who were not involved in patient’s care, based on the 1990 American College of Rheumatology (ACR) criteria (including TAB) TABLE 1. Different types of visual impairments reported in the HORTIM cohort at baseline Afferent system impairments related to 1 vascular network Isolated efferent system impairment Afferent visual impairments related to multiple vascular networks Mixed visual impairments Ophthalmic Diagnosis Vascular Network Involved AION CLRAO CRAO TVL ONP, either transient or fixed Ciliary arteries Ciliary arteries Central retinal artery Ciliary or central retinal arteries Ocular motor nerve arteries (ophtalmic artery system) Ciliary arteries with or without central retinal artery Ciliary and retinal central arteries Ciliary arteries Ocular motor nerve arteries associated with ciliary or central retinal arteries AION and previous TVL AION and CRAO AION and CLRAO ONP and AION ONP and CRAO ONP and AION and CRAO AION, anterior ischemic optic neuropathy; CLRAO, cilioretinal artery occlusion; CRAO, central retinal artery occlusion; ONP, oculomotor nerve palsy; TVL, transient visual loss. Dentel et al: J Neuro-Ophthalmol 2022; 42: 218-225 219 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 1. Sample images of RA illustrating a choroidal filling delay. A. FA (left) and ICG (right) angiograms in a right eye, 30 seconds after the dyes were injected: 2 typical triangle-shaped choroidal delay in the upper temporal territory (corresponding to the short posterior superior ciliary arteries), seen as well with both dyes (arrowheads). This patient was diagnosed with GCA and suffered from a contralateral anterior ischemic optic neuropathy. B. FA (left) and ICG (right) angiograms in a left eye, 40 seconds after the dyes were injected: typical triangle-shaped choroidal delay in nasal territory (corresponding to the short posterior nasal ciliary arteries), better on the indocyanine green angiography (arrowheads). This patient was diagnosed with GCA and suffered only from diplopia related to a left fourth nerve oculomotor palsy (repeated comprehensive ophthalmological examinations including visual fields excluded any optical or retinal ischemia). FA, fluorescein angiography; GCA, giant cell arteritis; ICG, indocyanine green angiography; RA, retinal angiography. (40). TAB was considered as positive or negative depending on the pathologist’s report. Analysis and Statistics FA, ICG, and MRI were compared with each other using the McNemar test for sensitivity and specificity and a generalized estimating equation test for a positive predictive value and negative predictive value. Positive and negative rates of each examination were studied according to the final diagnosis and the types of visual impairment. Finally, 4 imaging-based diagnostic approaches were studied: FA followed by ICG, ICG followed by FA, FA followed by MRI, and ICG followed by MRI. Confidence intervals were set at 95% and were calculated using the Clopper–Pearson method because the normal approximation might not be suitable because of the small amount of data. Chi-square or Fisher exact tests were used to compare categorical variables. Comparison of numerical vari220 ables was performed using the t test or Mann–Whitney U test as appropriate. All analyses were performed using R version 4.0.2. A P value , 0.05 was considered statistically significant. Ethics This retrospective study adhered to the tenants of the Declaration of Helsinki and has been approved by a research ethics board (IRB00012801). This study follows the Standards for Reporting of Diagnostic Accuracy Studies guidelines. RESULTS Forty-five patients were enrolled in the HORTIM cohort. Four patients were excluded, including 2 because of no visual impairment on admission, 1 because no MRI was available, and 1 because RA was not undergone. Among the Dentel et al: J Neuro-Ophthalmol 2022; 42: 218-225 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 2. Sample sagittal MRI section showing thickening and parietal contrast enhancement of the external carotid and the superficial temporal arteries (arrows). 41 patients included, 25 received a final diagnosis of GCA (Fig. 3). Patients’ characteristics are presented in Table 2. GCA patients were significantly older, more likely to have higher CRP and ESR rates, or to complain about systemic GCA symptoms. All GCA patients had systemic symptoms. Two of them only reported asthenia. Multiple afferent visual impairments and mixed visual impairments were most common in GCA patients, and the visual acuity in the affected eye was lower (1 vs 0.2 logMAR). In the nonGCA group, only 1 patient had a possible multiple afferent impairment involving retinal and ciliary networks. All patients who presented with retinal ischemia and oculomotor nerve palsy at baseline were diagnosed with GCA. RA was positive in 22 and 19 GCA cases (of 25) and negative in 11 and 13 non-GCA cases (of 16), respectively. The sensitivities of FA and ICG were 0.88 (0.69–0.97) and 0.76 (0.55–0.91), and the specificities were 0.69 (0.41– 0.89) and 0.81 (0.54–0.96) for the diagnosis of GCA, respectively. CE-VW MRI was positive in 23 of 25 GCA patients, which is a sensitivity of 0.92 (0.74–0.99) and a specificity of 0.94 (0.70–1.00). When compared statistically, there was no significant difference in sensitivity and specificity for the diagnosis of GCA between FA, ICG, and CE-VW MRI. Indeed, when comparing sensitivity of FA and ICG, the McNemar test did not show significant difference (P = 0.08). The same was found for specificity (P = 0.16). In addition, no significant difference was found between ICG and CE-VW MRI (P = 0.16 for sensitivity and P = 0.32 for specificity) and between FA and CE-VW MRI (P = 0.65 for sensitivity and P = 0.10 for specificity). However, absolute values showed the superiority of CE-VW MRI for sensitivity and specificity. The results of the different examinations according to the type of visual impairment are presented in Tables 3 and 4. RA and CE-VW MRI were positive in every GCA patient presenting multiple afferent visual impairments (n = 5). The 3 examinations were also negative in every non-GCA patient presenting with isolated diplopia (n = 4). By contrast, in the multiple afferent visual impairments group, 1 non-GCA patient showed choroidal delay on RA. In GCA patients complaining about isolated diplopia at baseline, 1 presented choroidal ischemia on RA. Finally, only 1 cilioretinal artery occlusion was found in addition to choroidal delay on FA, in a GCA patient. The different diagnostic approaches are presented in Table 5. RA’s specificity and sensitivity were not FIG. 3. Flowchart. Dentel et al: J Neuro-Ophthalmol 2022; 42: 218-225 221 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 2. Patients’ characteristics Patients’ Characteristics Final diagnosis of GCA (ACR 1990) Gender Female Male Mean age (extreme values) Baseline blood tests Mean CRP (extreme values) Mean ESR (extreme values) Presence of at least 1 symptom below Presence of at least 2 symptoms below Recent headaches Jaw claudication Scalp tenderness Fever Shoulders or hips pain General symptoms below Recent loss of body weight .10% Anorexia Asthenia Visual impairment with normal fundus§ Initial visual acuity of the impaired eye in logMAR (extreme values) Simple afferent visual impairment Isolated TVL Isolated AION Isolated CRAO Combined afferent visual impairment TVL + AION CLRAO + AION CRAO + AION Isolated efferent impairment (diplopia) Mixed visual impairment Diplopia + AION Diplopia + CRAO Diplopia + AION + CRAO Positive FA CLRAO + choroidal filling delay Positive ICG Positive CE-VW MRI (certain) Negative CE-VW MRI Scored suspicious Scored negative Number of Patients (%) n = 41 (100%) P value Yes n = 25 (61,0) No n = 16 (39,0) 16 (64,0) 9 (36,0) 79.4 (59–94) 7 (43,8) 9 (56,2) 71.4 (60–89) 0.20 70.0 (9–179)* 67.1 (15–114)† 23 (92,0) 18 (72,0) 18 (72,0) 16 (64.0) 12 (48,0) 0 (0.0) 2 (8,0) 15 (60,0) 2 (8,0) 1 (4,0) 15 (60,0) 0 (0.0) 1 (0–2.6) [0.15–2.3] 12.5 (1–36) 42.6 (3–100)‡ 8 (50.0) 1 (6.3) 7 (43,8) 0 (0.0) 2 (12,5) 0 (0.0) 0 (0.0) 6 (37,5) 0 (0.0) 1 (6,25) 6 (37,5) 1 (0.6) 0.2 (0–2.3) [0–0.6] ,0.001 0.04 0.007 ,0.001 13 (52,0) 2 (8,0) 10 (40,0) 1 (4,0) 5 (20,0) 2 (8,0) 1 (4,0) 2 (8,0) 2 (8,0) 4 (16,0) 1 (4,0) 1 (4,0) 2 (8,0) 22 (88,0) 1 (4,0) 19 (76,0) 23 (92,0) 2 (8,0) 2 (8,0) 0 (0.0) 9 (56,3) 2 (12,5) 7 (43,8) 0 (0.0) 2 (12,5) 2 (12,5) 0 (0.0) 0 (0.0) 4 (25,0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 5 (31,3) 0 (0.0) 3 (18,8) 1 (6,25) 15 (93,8) 1 (6,25) 14 (87,5) 0.006 0.16 1 0.026 0,80 0.68 0.19 0.14 ,0.001 ,0.001 ,0.001 ,0.001 *One value not available. † Five values not available. ‡ Three values not available. § Related to a cataract. AION, anterior ischemic optic neuropathy; ACR, American College of Rheumatology; CE-VW MRI, contrast-enhanced vessel-wall MRI; CLRAO, cilioretinal artery occlusion; CRP, C-reactive protein; CRAO, central retinal artery occlusion; ESR, erythrocyte sedimentation rate; FA, fluorescein angiography; GCA, giant cell arteritis; ICG, indocyanine green angiography; TVL, transient visual loss. significantly different, regardless of the number of dyes used and of the sequence (FA then ICG or ICG then FA). The approaches consisting in performing initially RA followed by MRI provide a sensitivity of 100% (P = 0.014, when compared with using only ICG) when ICG is negative and a specificity of 100% (P = 0.025, when compared with using only FA) when FA is positive. 222 CONCLUSIONS We found diagnostic accuracies of RA and MRI to be statistically similar in the diagnosis of cranial GCA revealed by visual signs. However, this result might be related to a lack of power, for we found a trend in favor of MRI, with a sensitivity reaching 92% and a specificity of Dentel et al: J Neuro-Ophthalmol 2022; 42: 218-225 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 3. Visual impairment in GCA patients in comparison with positivity of FA, ICG, and CE-VW MRI Visual impairment FA + (%) ICG + (%) CE-VW MRI + (%) Isolated Isolated Isolated Isolated Multiple Mixed 8 (80,0) 0 (0.0) 1 (100,0) 1 (50,0)* 5 (100,0) 2 (50,0) AION TVL CRAO diplopia afferent 9 (90,0) 2 (100,0) 1 (100,0) 1 (50,0)* 5 (100,0) 3 (75,0) 10 (100,0) 2 (100,0) 0 (0.0) 2 (100,0) 5 (100,0) 4 (100,0) *Same patient. AION, anterior ischemic optic neuropathy; CE-VW MRI, contrastenhanced vessel-wall MRI; CRAO, central retinal artery occlusion; FA, fluorescein angiography; ICG, indocyanine green angiography; TVL, transient visual loss. 94%. These findings are consistent with previously published data. Dejaco et al found sensitivities and specificities of 93% and 81% compared with TAB, respectively, based on studies with 2D MRI (23). Our study used a new process based on 3D MRI, used by Poillon et al who found better diagnostic accuracies of 3D vs 2D MRI, with sensitivities of 80% vs 70% and specificities of 100% vs 85% (34). In our study, MRI caught up every false negatives and false positives of RA, still allowing RA to confirm diagnosis in cases where MRI is considered as “suspicious” (concerned 2 patients in our study). Moreover, it is important to keep in mind that MRI presents several interpretation traps (temporal artery atherosclerosis for instance) that can lead to potential misinterpretation and false positives. Regardless of the type of dye used, RA reached good diagnostic accuracy in our cohort. In specific clinical circumstances, it performed as well as MRI. In all patients with multiple afferent visual impairments, including anterior ischemic optic neuropathy (AION) preceded by transient visual loss (TVL), RA had remarkable sensitivity. We noted that AF and ICG are misreading in same patients, which means there would be no added value to perform both type TABLE 4. Visual impairment in non-GCA patients in comparison with negativity of FA, ICG, and CE-VW MRI Visual Impairment Isolated Isolated Isolated Isolated Multiple Mixed AION TVL CRAO diplopia afferent FA (%) ICG (%) CE-VW MRI (%) 4 (57,1) 1 (50,0) † 4 (100,0) 1 (50,0)* † 5 (71,4) 2 (100,0) † 4 (100,0) 1 (50,0)* † 6 (85,7) 2 (100,0) † 4 (100,0) 2 (100,0) † *Same patient. † None of the patients filled these categories. AION, anterior ischemic optic neuropathy; CE-VW MRI, contrastenhanced vessel-wall MRI; CRAO, central retinal artery occlusion; FA, fluorescein angiography; ICG, indocyanine green angiography; TVL, transient visual loss. Dentel et al: J Neuro-Ophthalmol 2022; 42: 218-225 of RA. In addition, ICG tends to be more specific, whereas FA tends to be more sensitive. These findings should be considered regarding the adverse events fluorescein can induce, which can lead to death. ICG is far safer than fluorescein (41,42). We therefore recommend performing ICG rather than FA in the early management of GCA revealed by visual signs. However, RA presented too high false-negative and falsepositive rates to be considered as a standalone diagnostic tool. We believe that patients with visual manifestations and suspected GCA should undergo at least 2 imaging tests. We explored different approaches based on patients’ usual pathway through the health care system. Because very few centers are equipped with technical facilities that allow urgent MRI examinations, RA would be the examination obtained in the shortest delay. It seems that the best approach is to perform RA followed by MRI (Fig. 4). Considering patients with all types of visual impairments, performing ICG followed by MRI provides a sensitivity of 100% if ICG is mistaken and a specificity of 100% if not. This strategy allowed us to reach the correct diagnosis in all our patients. In this study, all the patients had TAB and 6 GCA patients had a mistaken TAB. The lack of sensitivity of TAB is well-known (5), and the European League Against Rheumatism society allowed the diagnosis of GCA without carrying TAB (22). Our strategy is consistent with this approach and ensures that the final assessment is much more accurate than TAB. This study presents several limitations, including a lack of power related to the number of included patients as discussed above. Considering MRI as negative if scored “suspicious” may also have minimized the chance of finding significant difference with RA. The retrospective design of this study introduces bias in data. However, inclusion in the HORTIM cohort and procedures/ examinations did occur in a prospective way, which may prevent retrospective bias. Although our algorithm is designed for use in primary care medicine, it was based on the findings of this study conducted in a tertiary center. However, only 5 patients (presenting isolated AIONs) of the 41 analyzed were referred by a physician. All the others primarily sought medical attention in our eye emergency department. Because this study was conducted in a tertiary center, MRI and RA were performed and interpreted by GCA experts. It should be kept in mind that vessel-wall MRI currently requires reading by experienced neuroradiologists who are familiar with interpretation pitfalls. RA must also be interpreted by ophthalmologists trained to identify choroidal delay. Finally, we chose to classify patients according to the type of visual impairment and to consider TVL as a constituent of a multiple afferent mechanism. This decision was based on the assumption that the pathophysiological mechanism of TVL cannot be precisely determined. 223 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 5. Diagnostic approaches and values according to the performed order of the different exams Approaches FA ICG (vs FA only) First Examination Negative FA Positive FA ICG FA (vs ICG only) Negative ICG Positive ICG FA CE-VW MRI (vs FA only) Negative FA Positive FA ICG CE-VW MRI (vs ICG only) Negative ICG Positive ICG After Second Examination se 88% sp 69% se 76% sp 81% se 88% sp 69% se 76% sp 81% se 100% sp 63% se 80% sp 100% se 100% sp 75% se 68% sp 100% [69–97] [41–89] [55–91] [54–96] [69–97] [41–89] [55–91] [54–96] [86–100] [35–85] [59–93] [79–100] [86–100] [48–93] [46–85] [79–100] P NA NA 0.83 0.16 0.08 0.16 NA NA 0.32 0.32 0.16 0.025 0.014 0.32 0.16 0.08 Sensitivity (se) and specificity (sp) reached after the second examination were compared with the sensitivity and the specificity of the first examination to determine the added value of the second examination. Statistically significant results are shown in bold. CE-VW MRI, contrast-enhanced vessel-wall MRI; FA, fluorescein angiography; ICG, indocyanine angiography. Given the presence of simultaneous AION and central retinal artery occlusion (CRAO) in the GCA group, we can consider TVL as a pre-CRAO or pre-AION state, regardless of what patients finally presented. This choice may have had a limited impact on our results because the number of TVL followed by AION is comparable in the 2 groups. To conclude, GCA revealed by visual signs is a challenging and urgent diagnosis. We believe that ICG is still an important front-line test, ideally supplemented by MRI, which are safe and highly effective examinations in GCA revealed by visual signs. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: A. Dentel, R. Hage, and T. Senè; b. Acquisition of data: G. Clavel, A. Lecler, F. Charbonneau, and J. Savatovsky; c. Analysis and interpretation of data: K. Zuber. Category 2: a. Drafting the manuscript: A. Dentel and R. Hage; b. Revising it for intellectual content: A. Dentel and R. Hage, T. Senè, G. Clavel, A. Lecler, F. Charbonneau, J. Savatovsky, and C. Vignal. Category 3: a. Final approval of the completed manuscript. A. Dentel, R. Hage, T. Senè, G. Clavel, A. Lecler, F. Charbonneau, J. Savatovsky, and C. Vignal. REFERENCES FIG. 4. Diagnostic algorithm based on retinal angiography (RA) as the first-line examination and contrast-enhanced vessel-wall MRI (CE-VW MRI) as the final examination in suspected giant cell arteritis (GCA) revealed by visual signs. AION, anterior ischemic optic neuropathy; CE-VW MRI, contrast-enhanced vessel-wall MRI; ICG, indocyanine green. 224 1. Wilkinson IM, Russell RW. Arteries of the head and neck in giant cell arteritis. A pathological study to show the pattern of arterial involvement. Arch Neurol. 1972;27:378–391. 2. Hunder GG. Epidemiology of giant-cell arteritis. Cleve Clin J Med. 2002;69:79–82. 3. Salvarani C, Cantini F, Boiardi L, Hunder GG. Polymyalgia rheumatica and giant-cell arteritis. N Engl J Med. 2002;347:261–271. 4. Rahman W, Rahman FZ. Giant cell (temporal) arteritis: an overview and update. Surv Ophthalmol. 2005;50:415–428. 5. Lyons HS, Quick V, Sinclair AJ, Nagaraju S, Mollan SP. A new era for giant cell arteritis. Eye (Lond). 2020;34:1013–1026. 6. Rinden T, Miller E, Nasr R. Giant cell arteritis: an updated review of an old disease. Cleve Clin J Med. 2019;86:465–472. 7. Vodopivec I, Rizzo JF III. Ophthalmic manifestations of giant cell arteritis. Rheumatology (Oxford). 2018;57:ii63–ii72. 8. Soriano A, Muratore F, Pipitone N, Boiardi L, Cimino L, Salvarani C. Visual loss and other cranial ischaemic complications in giant cell arteritis. Nat Rev Rheumatol. 2017;13:476–484. 9. Hayreh SS, Podhajsky PA, Zimmerman B. Occult giant cell arteritis: ocular manifestations. Am J Ophthalmol. 1998;125:521–526. 10. Vodopivec I, Rizzo JF. Ophthalmic manifestations of giant cell arteritis. Rheumatology (Oxford). 2018;57:ii63–ii72. Dentel et al: J Neuro-Ophthalmol 2022; 42: 218-225 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution 11. Singh AG, Kermani TA, Crowson CS, Weyand CM, Matteson EL, Warrington KJ. Visual manifestations in giant cell arteritis: trend over 5 decades in a population-based cohort. J Rheumatol. 2015;42:309–315. 12. Danesh-Meyer H, Savino PJ, Gamble GG. Poor prognosis of visual outcome after visual loss from giant cell arteritis. Ophthalmology. 2005;112:1098–1103. 13. Liu GT, Glaser JS, Schatz NJ, Smith JL. Visual morbidity in giant cell arteritis. Clinical characteristics and prognosis for vision. Ophthalmology. 1994;101:1779–1785. 14. Hocevar A, Rotar Z, Jese R, Semrl SS, Pizem J, Hawlina M, Tomsic M. Do early diagnosis and glucocorticoid treatment decrease the risk of permanent visual loss and early relapses in giant cell arteritis: a prospective longitudinal study. Medicine (Baltimore). 2016;95:e3210. 15. Hayreh SS, Zimmerman B, Khardon RH. Visual improvement with corticosteroid therapy in giant cell arteritis. Report of large study and review of literature. Acta Ophthalmol Scand. 2002;80:355–367. 16. González-Gay MA, Blanco R, Rodríguez-Valverde V, MartínezTaboada VM, Delgado-Rodriguez M, Figueroa M, Uriarte E. Permanent visual loss and cerebrovascular accidents in giant cell arteritis: predictors and response to treatment. Arthritis Rheum. 1998;41:1497–1504. 17. Aiello PD, Trautmann JC, McPhee TJ, Kunselman AR, Hunder GG. Visual prognosis in giant cell arteritis. Ophthalmology. 1993;100:550–555. 18. Liozon F, Catanzano G. Horton’s temporal arteritis. Anatomopathologic study using light microscopy. Apropos of 123 temporal biopsies. Rev Med Interne. 1982;3:295–301. 19. Quinn EM, Kearney DE, Kelly J, Keohane C, Redmond HP. Temporal artery biopsy is not required in all cases of suspected giant cell arteritis. Ann Vasc Surg. 2012;26:649–654. 20. Bowling K, Rait J, Atkinson J, Srinivas G. Temporal artery biopsy in the diagnosis of giant cell arteritis: does the end justify the means?. Ann Med Surg. 2017;20:1–5. 21. Sait MR, Lepore M, Kwasnicki R, Allington J, Srinivasaiah N. Temporal artery biopsy: time matters. Intern Med J. 2017;47:1465. 22. Hellmich B, Agueda A, Monti S, Buttgereit F, de Boysson H, Brouwer E, Cassie R, Cid MC, Dasgupta B, Dejaco C, Hatemi G, Hollinger N, Mahr A, Mollan SP, Mukhtyar C, Ponte C, Salvarani C, Sivakumar R, Tian X, Tomasson G, Turesson C, Schmidt W, Villiger PM, Watts R, Young C, Luqmani RA. Update of the EULAR recommendations for the management of large vessel vasculitis. Ann Rheum Dis. 2020;79:19–30. 23. Dejaco C, Ramiro S, Duftner C, Besson FL, Bley TA, Blockmans D, Brouwer E, Cimmino MA, Clark E, Dasgupta B, Diamantopoulos AP, Direskeneli H, Iagnocco A, Klink T, Neill L, Ponte C, Salvarani C, Slart RHJA, Whitlock M, Schmidt WA. EULAR recommendations for the use of imaging in large vessel vasculitis in clinical practice. Ann Rheum Dis. 2018;77:636– 643. 24. Klink T, Geiger J, Both M, Ness T, Heinzelmann S, Reinhard M, Holl-Ulrich K, Duwendag D, Vaith P, Bley TA. Giant cell arteritis: diagnostic accuracy of MR imaging of superficial cranial arteries in initial diagnosis-results from a multicenter trial. Radiology. 2014;273:844–852. 25. Bley TA, Wieben O, Vaith P, Schmidt D, Ghanem NA, Langer M. Magnetic resonance imaging depicts mural inflammation of the temporal artery in giant cell arteritis. Arthritis Rheum. 2004;51:1062–1064. 26. Bley TA, Wieben O, Uhl M, Thiel J, Schmidt D, Langer M. Highresolution MRI in giant cell arteritis: imaging of the wall of the superficial temporal artery. AJR Am J Roentgenol. 2005;184:283–287. 27. Bley TA, Uhl M, Carew J, Markl M, Schmidt D, Peter HH, Langer M, Wieben O. Diagnostic value of high-resolution MR imaging in Dentel et al: J Neuro-Ophthalmol 2022; 42: 218-225 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. giant cell arteritis. AJNR Am J Neuroradiol. 2007;28:1722– 1727. Bley TA, Reinhard M, Hauenstein C, Markl M, Warnatz K, Hetzel A, Uhl M, Vaith P, Langer M. Comparison of duplex sonography and high-resolution magnetic resonance imaging in the diagnosis of giant cell (temporal) arteritis. Arthritis Rheum. 2008;58:2574–2578. Rhéaume M, Rebello R, Pagnoux C, Carette S, Clements-Baker M, Cohen-Hallaleh V, Doucette-Preville D, Stanley Jackson B, Salama Sargious Salama S, Ioannidis G, Khalidi NA. Highresolution magnetic resonance imaging of scalp arteries for the diagnosis of giant cell arteritis: results of a prospective cohort study. Arthritis Rheumatol. 2017;69:161–168. Siemonsen S, Brekenfeld C, Holst B, Kaufmann-Buehler AK, Fiehler J, Bley TA. 3T MRI reveals extra- and intracranial involvement in giant cell arteritis. AJNR Am J Neuroradiol. 2015;36:91–97. Epub 2014 Aug 28. Tedeschi SK, Aghayev A. Picturing giant cell arteritis: projecting into the future. Arthritis Rheumatol. 2019;71:1211–1214. Sommer NN, Treitl KM, Coppenrath E, Kooijman H, Dechant C, Czihal M, Kolben TM, Beyer SE, Sommer WH, Saam T. ThreeDimensional high-resolution black-blood magnetic resonance imaging for detection of arteritic anterior ischemic optic neuropathy in patients with giant cell arteritis. Invest Radiol. 2018;53:698–704. Mohammed-Brahim N, Clavel G, Charbonneau F, Duron L, Picard H, Zuber K, Savatovsky J, Lecler A. Three Tesla 3D highresolution vessel wall MRI of the orbit may differentiate arteritic from nonarteritic anterior ischemic optic neuropathy. Invest Radiol. 2019;54:712–718. Poillon G, Collin A, Benhamou Y, Clavel G, Savatovsky J, Pinson C, Zuber K, Charbonneau F, Vignal C, Picard H, Leturcq T, Miranda S, Sené T, Gerardin E, Lecler A. Increased diagnostic accuracy of giant cell arteritis using threedimensional fat-saturated contrast-enhanced vessel-wall magnetic resonance imaging at 3 T. Eur Radiol. 2020;30:1866–1875. Siatkowski RM, Gass JD, Glaser JS, Smith JL, Schatz NJ, Schiffman J. Fluorescein angiography in the diagnosis of giant cell arteritis. Am J Ophthalmol. 1993;115:57–63. Hayreh SS. Posterior ciliary artery occlusion. Ophthalmol Retina. 2018;2:106–111. Pellegrini M, Giannaccare G, Bernabei F, Moscardelli F, Schiavi C, Campos EC. Choroidal vascular changes in arteritic and nonarteritic anterior ischemic optic neuropathy. Am J Ophthalmol. 2019;205:43–49. Mandell DM, Mossa-Basha M, Qiao Y, Hess CP, Hui F, Matouk C, Johnson MH, Daemen MJ, Vossough A, Edjlali M, Saloner D, Ansari SA, Wasserman BA, Mikulis DJ. Vessel wall imaging study group of the American society of neuroradiology. Intracranial vessel wall MRI: principles and expert consensus recommendations of the American society of neuroradiology. AJNR Am J Neuroradiol. 2017;38:218–229. Bei L, Lee I, Lee MS, Van Stavern GP, McClelland CM. Acute vision loss and choroidal filling delay in the absence of giantcell arteritis. Clin Ophthalmol. 2016;10:1573–1578. Hunder GG, Bloch DA, Michel BA, Stevens MB, Arend WP, Calabrese LH, Edworthy SM, Fauci AS, Leavitt RY, Lie JT. The American College of Rheumatology 1990 criteria for the classification of giant cell arteritis. Arthritis Rheum. 1990;33:1122–1128. Kwiterovich KA, Maguire MG, Murphy RP, Schachat AP, Bressler NM, Bressler SB, Fine SL. Frequency of adverse systemic reactions after fluorescein angiography. Results of a prospective study. Ophthalmology. 1991;98:1139–1142. Hope-Ross M, Yannuzzi LA, Gragoudas ES, Guyer DR, Slakter JS, Sorenson JA, Krupsky S, Orlock DA, Puliafito CA. Adverse reactions due to indocyanine green. Ophthalmology. 1994;101:529–533. 225 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.