Fluorescein angiography (FA) was first described in the 1960s as a method for visualizing the retinal and choroidal vessels.1,2 FA involves injection of fluorescein dye into a peripheral vein followed by imaging of the retina for up to 10 minutes. This technique has been the gold standard for identifying or evaluating a variety of retinal vascular diseases. It is particularly useful for visualizing vascular filling defects, areas of capillary nonperfusion, and retinal vessel tortuosity, dilation, aneurysms, and leakage.3 FA is an important prognostic indicator in retinal vascular diseases such as central and branch retinal vein occlusions in which areas of nonperfusion are indicative of future visual acuity.4,5 FA is also useful in diabetic retinopathy to detect vascular leakage, capillary nonperfusion, and retinal neovascularization.6
However, FA has a few drawbacks that limit its routine use. Because it is expensive and can take up to 10 minutes, it is not an ideal technique to use on a regular basis in a busy clinical setting. The technique is also invasive, requiring intravenous injection of dye. Although comparatively safe, the dye is relatively contraindicated in patients who are pregnant or who have severe renal disease, and poses risks ranging from nausea to allergic reactions, including anaphylaxis in rare instances.7,8 For these reasons, a rapid noninvasive technique to visualize retinal and choroidal vessels would be beneficial for the evaluation of patients requiring frequent follow-up examinations or those who may not tolerate injection of intravenous dye.
Optical coherence tomography angiography (OCTA) is a new imaging technique that may afford clinicians this advantage. OCTA maps erythrocyte flow by comparing the decorrelation signal between sequential OCT B-scans taken at the same cross-section. Axial bulk motion from patient movement is removed so locations of motion between repeated OCT B-scans signify erythrocyte movement over time and, therefore, vessels. OCTA is a noninvasive imaging modality that creates high-resolution volumetric OCTA in a matter of seconds.9–12
The major disadvantage of OCTA is a small field of view, which is limited by scanning speed. For a given acquisition time, there is a trade-off between larger field of view and lower image quality, due to under-sampling (Figure 1). In clinical practice, a wide field of view of the retinal vasculature is often critical to visualize vascular abnormalities in a variety of retinal disease states. Therefore, techniques to widen the field of view of OCTA are clinically important. This study explores a method of combining multiple 3 × 3–mm OCTA images to create wide-field montage OCTA images to improve visualization of retinal vasculature in the posterior pole.
Comparison of 3 × 3–mm, 6 × 6–mm, and 8 × 8–mm OCT angiography (OCTA) images from a normal eye using the prototype AngioVue OCTA software on the RTVue XR Avanti spectral-domain OCT (SD-OCT) device (Optovue, Fremont, CA). (A) 3 × 3–mm OCTA showing a detailed view of the vasculature. (B) 6 × 6–mm OCTA. (C) 8 × 8–mm OCTA demonstrating a wide view of the posterior pole but decreased visualization of retinal vascular details.
Patients and Methods
This study was approved by the Tufts Medical Center Institutional Review Board, and informed consent was obtained from patients prior to examination in accordance with the Tufts Medical Center Institutional Review Board. The research adhered to the tenets of the Declaration of Helsinki and complied with the Health Insurance Portability and Accountability Act of 1996.
In this prospective study, patients were imaged using the prototype AngioVue OCTA software on the RTVue XR Avanti spectral-domain OCT (SD-OCT) device (Optovue, Fremont, CA) in November 2014 at the New England Eye Center at Tufts Medical Center. The AngioVue OCTA software used a split-spectrum amplitude decorrelation angiography algorithm to improve the signal-to-noise ratio. The Avanti device operated at 70,000 A-scans per second to acquire OCTA volumes consisting of 304 × 304 A-scans in approximately 2.6 seconds. Each OCTA was created using orthogonal registration and merging of two consecutive scan volumes. Nine 3 × 3–mm OCTA images were obtained in adjacent regions of the posterior pole by moving the software’s scanning area without moving fixation. Adobe Photoshop (Adobe Systems, San Jose, CA) then was used to manually piece together the 3 × 3–mm OCTA images to create a single wide-field montage OCTA image of approximately 8 × 8–mm or a 30° field. For comparison, a single scan 8 × 8–mm OCTA image was obtained using the same amount of A-scans (304 × 304) as the 3 × 3–mm scans. Fifty-degree FA images also were acquired in all patients using the Spectralis (Heidelberg Engineering, Heidelberg, Germany). The FA images were cropped to match the field of the single-scan 8 × 8–mm OCTA images, which is approximately equivalent to a 30° field.
A total of five eyes of three patients were imaged for this study: one patient with two healthy eyes, one patient with a branch retinal vein occlusion in the right eye and a healthy left eye, and one patient with bilateral severe proliferative diabetic retinopathy whose right eye was imaged. The five eyes were evaluated using montage OCTA, 8 × 8–mm OCTA, and FA. The montage OCTA showed the retinal vasculature in greatest detail. The montage OCTA technique successfully detected vascular abnormalities seen on FA, and in some instances, pathology that was not detected by FA and 8 × 8–mm OCTA was visualized.
A 56-year-old man was referred for evaluation of intermittent uveitis in his left eye. He had a history of biopsy-proven sarcoidosis as well as psoriatic arthritis, which was controlled with subcutaneous adalimumab (Humira; AbbVie, North Chicago, IL). Visual acuity was 20/20-2 in the right eye and 20/25-2 in the left eye. Dilated fundus examination and FA were normal and without signs of active uveitis. A wide-field montage OCTA image of each eye was created, demonstrating the appearance of the normal retinal vasculature in the posterior pole. Wide-field montage OCTA images showed the retinal capillaries in much greater detail compared with the FA and single-scan 8 × 8–mm OCTA images (Figures 2 and 3).
Case 1. The right eye of a 56-year-old man with a normal retina. (A) Fluorescein angiography appears normal, depicting the larger retinal vessels. (B) The wide-field montage OCT angiography (OCTA) image clearly delineates the normal retinal vasculature including the smaller capillaries. (C) A single-scan 8 × 8–mm OCTA image shows the larger retinal vessels and some of the smaller capillaries but detailed visualization of the capillary plexus is not feasible.
Case 1. The left eye of a 56-year-old man with a normal retina. (A) Fluorescein angiography appears normal, showing the larger retinal vessels. (B) The wide-field montage OCT angiography (OCTA) image clearly delineates the normal retinal vasculature including the smaller capillaries. (C) A single-scan 8 × 8–mm OCTA image shows the larger retinal vessels and some of the smaller capillaries but detailed visualization of the capillary plexus is not feasible.
A 53-year-old woman with a history of hypertension presented with a branch retinal vein occlusion in the right eye with associated macular edema and severe peripheral nonperfusion. She received 10 intravitreal bevacizumab injections, as well as sectoral panretinal photocoagulation laser treatment (PRP). Her retinopathy had stabilized, and on follow-up examination, visual acuity was 20/20-3 in the right eye and 20/20-1 in the left eye. The dilated fundus examination showed an inferior branch retinal vein occlusion with sclerotic vessels and few exudates in the superior macula without cystoid macular edema or neovascularization, and inferior sectoral PRP scars in the right eye. Few calcific drusen were visualized on fundus examination in the left eye. Both FA and OCTA were performed, and a wide-field montage OCTA image was created.
The FA and wide-field montage OCTA images of the left eye demonstrated fine telangiectatic vessels along the inferior arcade, possibly secondary to a previous subclinical small vein occlusion. The wide-field montage OCTA image of the left eye also demonstrated mild capillary nonperfusion in the area of the abnormal vessels, which was not apparent on FA or the single-scan 8 × 8–mm OCTA image (Figure 4).
Case 2. The normal left eye of a 53-year-old woman with a contralateral branch retinal vein occlusion. (A) Fluorescein angiography (FA) showing fine telangiectases along the inferior arcade. (B) The wide-field montage OCT angiography (OCTA) image demonstrates delicate telangiectatic vessels and capillary nonperfusion along the inferior arcade. (C) A single-scan 8 × 8–mm OCTA image appears grossly normal and is unable to visualize small vascular abnormalities detected on FA and wide-field montage OCTA.
The FA, single-scan 8 × 8–mm OCTA image, and wide-field montage OCTA image of the right eye all clearly demonstrated a large wedge-shaped area of capillary nonperfusion in the inferotemporal macula. The wide-field montage OCTA image more clearly delineated the boundary of nonperfusion, providing the ability to visualize the individual vascular abnormalities such as microaneurysms, telangiectases, and anastomoses. However, the 50° FA images better displayed areas of laser treatment and were capable of a larger field of view, which would have been useful for detecting peripheral neovascularization had there been any in this patient (Figure 5).
Case 2. A 53-year-old woman with a branch retinal vein occlusion in the right eye. A large, wedge-shaped area of capillary nonperfusion in the inferotemporal macula is evident. (A) Fluorescein angiography displaying telangiectatic vessels (yellow arrow) and microaneurysms (yellow circle) along the margin of ischemia and areas of prior laser treatment. Vascular detail is lost in the periphery of the image. (B) The wide-field montage OCT angiography (OCTA) image more clearly delineates the boundary of nonperfusion, providing an enhanced ability to visualize the individual vessels and therefore microaneurysms (yellow circle), telangiectases (yellow arrow), and anastomoses. (C) A single-scan 8 × 8–mm OCTA image shows the border of the capillary nonperfusion, but details of many of the microvascular abnormalities are not apparent.
A 31-year-old man with a history of type 1 diabetes mellitus with proliferative diabetic retinopathy stabilized with PRP in both eyes presented for a follow-up visit. Visual acuity was 20/30 in the right eye. Dilated fundus examination in the right eye showed small patches of neovascularization of the disc (NVD) and neovascularization elsewhere (NVE), absent diabetic macular edema, trace cotton wool spots, few scattered hard exudates, and a moderate amount of PRP scars in the midperipheral retina.
FA demonstrated late leakage from areas of NVD and early NVE. FA also revealed an enlarged foveal avascular zone (FAZ), multiple microaneurysms, and areas of capillary nonperfusion in the posterior pole. A single-scan 8 × 8–mm OCTA image also depicted larger areas of capillary nonperfusion, NVE, and an enlarged FAZ; however, it was difficult to detect microaneurysms using this image. The wide-field montage OCTA image showed the retinal vasculature in higher detail, including an enlarged FAZ and perifoveal intercapillary area, multiple microaneurysms, early NVE, and areas of capillary nonperfusion including areas too small to visualize on the FA or the single-scan 8 × 8–mm OCTA image (Figure 6).
Case 3. The right eye of a 31-year-old man with proliferative diabetic retinopathy. (A) Fluorescein angiography (FA) demonstrates early neovascularization elsewhere (NVE) (yellow circles), an enlarged foveal avascular zone (FAZ), multiple microaneurysms (red circles), and areas of capillary nonperfusion (yellow arrows). (B) The wide-field montage OCT angiography (OCTA) image demonstrates the retinal vasculature in higher detail, showing areas of capillary nonperfusion (yellow arrows) including areas too small to visualize on the FA or the single-scan 8 × 8–mm OCTA image, an enlarged FAZ and perifoveal intercapillary area, multiple microaneurysms highlighted by areas of capillary nonperfusion (red circles), telangiectases (red arrows), and early NVE (yellow circles). (C) A single-scan 8 × 8–mm OCTA image also depicts larger areas of capillary nonperfusion (yellow arrows), NVE (yellow circles), and an enlarged FAZ. Microaneurysms are difficult to visualize using this scan.
The wide-field montage OCTA technique is capable of capturing high-resolution images of the retinal vasculature using a noninvasive method. Similar montage techniques have been used to create wide-field fundus photographs, wide-field OCT B-scans, and other wide-field images.13,14 The technique of wide-field montage OCTA maintains the benefit of high-resolution imaging while addressing the limitation of a small field of view, which is one of the major drawbacks of OCTA. Wide-field montage OCTA enhances visualization of vessel pathology with a field of view of approximately 8 × 8–mm or 30° without moving the patient’s fixation point. The retinal periphery, although not demonstrated in this case series, could be included in the mosaic by adjusting the fixation location.
When evaluating retinal diseases, a large field of view with good detail of the retinal vasculature is important. FA is capable of wide-field imaging of the retinal vasculature and is used for verification of the diagnosis in a variety of retinal vascular diseases, including retinal vein occlusion and diabetic retinopathy.15 However, FA is invasive and is contraindicated in some patients; in addition, visualization of the retinal vasculature is limited in cases with retinal hemorrhage, scarring, or early leakage.4 Wide-field montage OCTA offers a noninvasive imaging modality that allows for improved visualization of the retinal vasculature. Figure 5 demonstrates the ability of montage OCTA to visualize the borders of capillary nonperfusion in detail and intricate vessel abnormalities such as microaneurysms and telangiectases.
In diabetic retinopathy, microaneuryms are typically the earliest sign of retinopathy, and FA has been shown to be sensitive in detecting these early findings. FA also has been a valuable tool for visualizing other vascular abnormalities, including capillary nonperfusion and retinal neovascularization.6 Wide-field montage OCTA images, as seen in Figure 6, can demonstrate these abnormalities and delineate subtle areas of capillary nonperfusion that are difficult to visualize using FA or single-scan 8 × 8–mm OCTA images.
Currently, the major limitation of the wide-field montage OCTA technique is the time it takes to obtain the completed image. The posterior pole can be constructed using nine individual 3 × 3–mm OCTA scans, each of which takes approximately one minute to acquire, including processing time. Manual stitching of the images via Adobe Photoshop then takes approximately 20 minutes. Incorporation of an automated montaging software would make the process faster and more feasible in a busy clinic setting.
Wide-field montage OCTA imaging enables high-resolution wide-field imaging of the retinal vasculature using a noninvasive imaging technique. Wide-field montage OCTA could provide the benefits of OCTA and also address the main drawback of OCTA, a small field of view. Automated montaging software would enhance the speed of performing this technique.
- Novotny HR, Alvis DL. A method of photographing fluorescence in circulating blood in the human retina. Circulation. 1961;24:82–86. doi:10.1161/01.CIR.24.1.82 [CrossRef]
- Novotny HR, Alvis D. A method of photographing fluorescence in circulating blood of the human eye. Tech Doc Rep SAMTDR USAF Sch Aerosp Med. 1960;60–82:1–4.
- Johnson RN, Fu AD, McDonald HR, et al. Fluorescein angiography: basic principles and interpretation. In: Ryan SJ, Sadda SR, Hinton DR, eds. Retina. 5th ed. London: Elsevier Saunders; 2013:2–50.
- Yuan A, Kaiser PK. Branch vein occlusion. In: Ryan SJ, Sadda SR, Hinton DR, eds. Retina. 5th ed. London: Elsevier Saunders; 2013:1029–1038.
- Laatikainen L, Kohner EM. Fluorescein angiography and its prognostic significance in central retinal vein occlusion. Br J Ophthalmol. 1976;60(6):411–418. doi:10.1136/bjo.60.6.411 [CrossRef]
- Zhang K, Ferreyra HA, Grob S, Bedell M, Zhang JJ. Diabetic retinopathy: genetic and etiologic mechanisms. In: Ryan SJ, Sadda SR, Hinton DR, eds. Retina. 5th ed. London: Elsevier Saunders; 2013:925–939.
- Kwiterovich KA, Maguire MG, Murphy RP, et al. Frequency of adverse systemic reactions after fluorescein angiography: results of a prospective study. Ophthalmology. 1991;98(7):1139–1142. doi:10.1016/S0161-6420(91)32165-1 [CrossRef]
- Lopez-Saez MP, Ordoqui E, Tornero P, et al. Fluorescein-induced allergic reaction. Ann Allergy Asthma Immunol. 1998;81(5 pt 1):428–430. doi:10.1016/S1081-1206(10)63140-7 [CrossRef]
- Kim DY, Fingler J, Zawadzki RJ, et al. Optical imaging of the chorioretinal vasculature in the living human eye. Proc Natl Acad Sci USA. 2013;110(35):14354–14359. doi:10.1073/pnas.1307315110 [CrossRef]
- Choi W, Mohler KJ, Potsaid B, et al. Choriocapillaris and choroidal microvasculature imaging with ultrahigh speed OCT angiography. PloS One. 2013;8(12):e81499. doi:10.1371/journal.pone.0081499 [CrossRef]
- Schwartz DM, Fingler J, Kim DY, et al. Phase-variance optical coherence tomography: a technique for noninvasive angiography. Ophthalmology. 2014;121(1):180–187. doi:10.1016/j.ophtha.2013.09.002 [CrossRef]
- Spaide RF, Klancnik JM Jr, Cooney MJ. Retinal vascular layers imaged by fluorescein angiography and optical coherence tomography angiography. JAMA Ophthalmol. 2015;133(1):45–50. doi:10.1001/jamaophthalmol.2014.3616 [CrossRef].
- Early Treatment Diabetic Retinopathy Study Research Group. Grading diabetic retinopathy from stereoscopic color fundus photographs—an extension of the modified Airlie House classification. ET-DRS report number 10. Ophthalmology. 1991;98(5)(suppl):786–806. doi:10.1016/S0161-6420(13)38012-9 [CrossRef]
- Mori K, Kanno J, Gehlbach PL, Yoneya S. Montage images of spectral-domain optical coherence tomography in eyes with idiopathic macular holes. Ophthalmology. 2012;119(12):2600–2608. doi:10.1016/j.ophtha.2012.06.027 [CrossRef]
- Ross Russell RW, Ffytche TJ, Sanders MD. A study of retinal vascular occlusion using fluorescein angiography. The Lancet. 1966;288(7468):821–825. doi:10.1016/S0140-6736(66)92254-9 [CrossRef]