Detection of choroidal hypoperfusion in giant cell arteritis using swept-source optical coherence tomographic angiography

Giant cell arteritis (GCA) is an immune-mediated vasculitis of medium to large sized arteries that predominantly affects adults over the age of 50 with high morbidity and mortality if untreated. GCA affects the short posterior ciliary arteries supplying the prelaminar and laminar portions of the optic nerve head, resulting in acute vision loss and pallid nerve edema characteristic of arteritic anterior ischemic optic neuropathy (AAION).¹ Other hallmark symptoms of GCA include headache, myalgias, fatigue, fever, weight loss, or jaw claudication. Fluorescein angiography (FA) classically demonstrates delayed or incomplete choroidal filling in the acute phase of GCA.² Optical coherence tomographic angiography (OCTA), a non-invasive tool for imaging of laminar blood flow, provides high-resolution, three-dimensional segmentation of the chorioretinal microvasculature into the superficial capillary plexus (SCP), deep capillary plexus (DCP) and choriocapillaris (CC).³ While FA has historically been the standard imaging modality for chorioretinal vascular pathology, OCTA is a promising alternative to dye-based angiography.

A woman in her late 70s with a history of rheumatoid arthritis presented with sudden vision loss in the right eye (OD) one day prior, with jaw claudication, headache, scalp tenderness and weight loss. On examination, visual acuities were counting fingers OD and 20/30 in the left eye (OS). There was a brisk relative afferent pupillary defect OD and a dense inferior arcuate defect OS on visual field testing (figure 1F; inset). Funduscopic exam revealed pallid optic disc edema that was circumferential in the right eye and most prominent superonasally in the left eye (figure 1A, F). Inflammatory markers and platelets (PLT) were elevated (ESR 119 mm/hour, CRP 75 mg/L, PLT 748 K/uL). Right temporal artery biopsy was positive for giant cell arteritis (GCA).

Figure 1.

A woman in her 70s with GCA causing sudden vision loss OD. Fundus images of the right (A) and left (F) eyes. Automated (Humphrey) perimetry (30-2) of the left eye (F; inset). FA images taken in the mid-phase (1:21) and early phase (0:25) for the right (B) and left (G) eyes, respectively. En face SS-OCTA 6×6 mm images with laminar segmentation of the superficial capillary plexus (C, H), deep capillary plexus (D, I) and choriocapillaris (E, J) for the right and left eyes, respectively (Triton, Topcon, Tokyo, Japan; off-label use under IRB-approved protocol). Insets depict B-scan images through the fovea with segmented area highlighted in blue. Red arrows outline matching boundaries of perfusion/nonperfusion on FA and SS-OCTA. Green arrows indicate decreased angiographic signal in the perfusion beds of small cilioretinal arteries.

Fluorescein angiography (FA) demonstrated a large, sharply demarcated area of choroidal filling delay involving the temporal fundus in the right eye (figure 1B). The left eye also demonstrated a large, sharply demarcated area of delayed choroidal perfusion centrally that extended inferiorly and temporally (figure 1G). Late frames showed full eventual choroidal filling in these areas and leakage at the optic nerve head in both eyes (not shown). Superficial capillary plexus (SCP) and deep capillary plexus (DCP) segmentation of en face swept-source optical coherence tomographic angiography (SS-OCTA) images demonstrated reduced angiographic signal in the perfusion beds of small cilioretinal arteries (figures 1C, H, I; green arrows). There was reduction in the SCP density temporally in the right eye, potentially signifying an early consequence of ganglion cell layer and retinal nerve fiber layer atrophy. Notably, segmentation of the choriocapillaris (CC) demonstrated decreased perfusion in the same distribution as the choroidal filling delay observed on FA (figure 1E, J). The patient was treated with intravenous methylprednisolone and transitioned to an oral prednisone taper. Follow-up exam 6 months later revealed visual acuities of no light perception OD and improvement to 20/25 OS with stable visual field defects.

For comparison, we present one patient without visual symptoms and biopsy-proven GCA was imaged with SS-OCTA as a control (figure 2). The patient was a woman in her 70s referred by her neurologist to evaluate for ophthalmic signs suggestive of GCA in the setting of elevated inflammatory markers (erythrocyte sedimentation rate (ESR) 65 mm/hour and C-reactive protein (CRP) 111 mg/L), jaw claudication (confounded by active dental disease), weight loss, and headaches with scalp tenderness. The patient denied visual changes. FA was obtained given her atypical clinical presentation, and no choroidal filling defects were found. Subsequent right temporal artery biopsy was positive, and the patient was treated with corticosteroids.

Figure 2.

Swept-source optical coherence tomographic angiography (SS-OCTA) images from a control, GCA-positive patient in her 70s without ocular involvement. Fundus images of the right (A) and left (F) eyes. Fluorescein angiography (FA) images taken in the early phase (0:17) and mid-phase (0:46) for the right (B) and left (G) eyes, respectively. En face SS-OCTA 6×6 mm images with laminar segmentation of the superficial capillary plexus (C, H), deep capillary plexus (D, I) and choriocapillaris (E, J) for the right and left eyes, respectively. Horizontal lines on SS-OCTA images are due to motion artifact.

Choroidal filling delay demonstrated by IV dye-based angiography is a highly suggestive finding of GCA in the right clinical context, due to hypoperfusion of the posterior ciliary arteries.² As such, FA serves as a useful ancillary tool in the evaluation of GCA, particularly in mild cases. OCTA has the potential to serve as a practical alternative to FA. FA testing has several limitations, including the need for a licensed practitioner to place an IV and administer fluorescein, the longer time required to complete the test, adverse reactions to fluorescein dye, and the inability to repeat FA testing in the event of poor image quality until the dye is excreted. On the other hand, OCTA is fast, easy to capture and non-invasive.³ Thus, OCTA may enable more expedient assessment of choroidal perfusion than dye-based angiography to facilitate the diagnosis and treatment of GCA in select cases.

We previously reported 4 GCA cases in which spectral-domain (SD)-OCTA demonstrated superficial peripapillary dilation and retinal capillary perfusion defects that corresponded to visual field loss in the acute phase.⁴ Two patients in the series had choroidal perfusion abnormalities evident on FA, yet SD-OCTA analysis of the choroid and CC (performed on the same day as FA) showed no signal abnormality in those corresponding regions. This case suggests SS-OCTA may offer improved resolution of deeper retinal and choroidal vessels as compared to SD-OCTA and thus might be better suited to capture alterations in choroidal perfusion.

Tran et al. reported the first definitive case of choroidal hypoperfusion secondary to GCA imaged using SS-OCTA to offer a direct comparison with FA and indocyanine green angiography.⁵ The case depicted a large Amalric choroidal infarct similar to the right eye in this case (figure 1A–E). Notably, Tran et al. employed a different SS-OCTA device (Plex-Elite 9000, Carl Zeiss Meditec, Dublin, CA) than the one used in our study and allowed for direct OCTA imaging of the full choroid. While their findings are similar to our findings, there are likely to be differences between SS-OCTA devices that influence sensitivity in detecting subretinal perfusion abnormalities.

This case demonstrates comparability between SS-OCTA and FA in detecting choroidal ischemia in GCA. There are many advantages of SS-OCTA over FA, including ease and expedience of obtaining imaging in the acute setting that is highly relevant in clinical scenarios of suspected GCA. Further study evaluating SS-OCTA as a diagnostic tool in screening for choroidal perfusion abnormalities in suspected GCA cases would be of great value.

Abbreviations:

SS-OCTA

swept-source optical coherence tomographic angiography

FA

fluorescein angiography

GCA

giant cell arteritis

SCP

superficial capillary plexus

DCP

deep capillary plexus

CC

choriocapillaris

SD

spectral-domain

PLT

platelet

ESR

erythrocyte sedimentation rate

CRP

C-reactive protein