To determine the intrascleral location of the circle of Zinn-Haller by simultaneous indocyanine green (ICG) angiography and enhanced-depth imaging optical coherence tomography (EDI-OCT) in highly myopic eyes.
Retrospective, consecutive, observational case series.
Ninety-four eyes of 67 consecutive patients with pathologic myopia who underwent simultaneous ICG angiography and EDI-OCT examinations by Spectralis HRA-OCT, and whose Zinn-Haller ring was observed within the area of myopic conus by ICG angiography, were studied. The definition of pathologic myopia was a refractive error (spherical equivalent) <−8.00 diopters (D) or an axial length >26.5 mm.
The EDI-OCT images showed cross-sections of the vessels that were identified in the ICG images as the circle of Zinn-Haller. The vessels were seen as a hyporeflective circle within the peripapillary sclera. An intrascleral course of the Zinn-Haller ring was clearly observed in adjacent serial OCT sections. The filling of the Zinn-Haller ring was from the short posterior ciliary arteries, and OCT also showed a continuous pathway from the retrobulbar short posterior ciliary arteries to the circle of Zinn-Haller. Centripetal branches were seen to run toward the optic nerve from the Zinn-Haller ring in 20 eyes by ICG and were confirmed by OCT in 4 eyes.
The HRA-OCT images confirmed that the vascular structure surrounding the optic disc observed by ICG angiography had topographic features specific to the Zinn-Haller arterial ring by OCT. The in situ observation of the circle of Zinn-Haller by simultaneous ICG angiography and OCT is a useful method to examine the Zinn-Haller ring in eyes with pathologic myopia.
The circle of Zinn-Haller is an intrascleral arteriolar anastomosis derived from the paraoptic medial and lateral short posterior ciliary arteries. The Zinn-Haller arterial ring is the main vascular supply of the optic nerve at the level of the lamina cribrosa, which is the major insult site of optic nerve damage in eyes with glaucoma. Because of its intrascleral location, it had been difficult to observe in situ. Thus, most of the studies of the circle of Zinn-Haller have used histologic sections or vascular castings with methyl methacrylate of human cadaver eyes. In a landmark study, Olver and associates examined ocular microvascular corrosion casts of human cadaver eyes and reported that the circle of Zinn-Haller was present in all of the cases. Recently, Jonas and Jonas examined histologic sections of 29 human eyes and found that the circle of Zinn-Haller was present in all of the eyes.
In earlier in situ studies, the Zinn-Haller arterial ring of human eyes was observed by fluorescein fundus angiography and indocyanine green (ICG) angiography in eyes with pathologic myopia with a large peripapillary atrophy. However, even with ICG angiography, it was still difficult to observe the entire course of the circle of Zinn-Haller in non–highly myopic eyes without a large peripapillary atrophy. Thus, Ko and associates examined the images of fluorescein angiography and were able to see the temporal part of the circle of Zinn-Haller in 15 eyes with pathologic myopia. More recently, we performed ICG angiography on 382 highly myopic eyes and found that the circle of Zinn-Haller was visible in 206 of 382 eyes (53.9%). In 162 of the 206 eyes, we were able to see only a part of the Zinn-Haller ring, but in the remaining 44 eyes, the circle of Zinn-Haller was seen to almost completely surround the optic nerve head. The blood to the circle of Zinn-Haller was seen to arise either from the lateral or medial short posterior ciliary arteries or from both short posterior ciliary arteries. Branches from the circle of Zinn-Haller were seen to run centripetally toward the optic nerve. Such angiographic examinations were advantageous because it was possible to delineate the dye filling pattern of the circle of Zinn-Haller. However, the images of fluorescein angiography and ICG angiography are 2-dimensional, and other important structural features of the circle of Zinn-Haller (eg, the location in the sclera) could not be observed.
In a study of the circle of Zinn-Haller in situ with methods other than angiography, Povazay and associates analyzed the vasculature of the optic nerve head with a 3-dimensional optical coherence tomography (3D OCT) detection system with a high-speed InGaAs camera (SU-LDH 1024; SUI Goodrich, Princeton, New Jersey, USA) at 1060 nm in 7 subjects. They showed that the circle of Zinn-Haller was visible beneath the choroidal–scleral border as a circular anastomosis surrounding the optic nerve head. This was an important study because the authors showed for the first time that the in situ circle of Zinn-Haller was situated in the sclera of human eyes. Unfortunately, the image quality was not good enough to provide more detailed information on the circle of Zinn-Haller.
The recently developed enhanced-depth imaging optical coherence tomography (EDI-OCT) has provided clear cross-sectional images of tissues deeper than the retina, such as the choroid and sclera. It was reported that the entire thickness of the sclera can be observed in highly myopic eyes with thin retinas and choroids. The Spectralis HRA/OCT instrument combines spectral-domain OCT (SD-OCT) with a confocal scanning laser ophthalmoscope (cSLO), thus allowing simultaneous recordings of cross-sectional OCT images and different topographic imaging methods such as ICG angiography. The OCT and ICG images are displayed side-by-side on the monitor screen. An exact alignment of the SD-OCT scans with the topographic cSLO images allows the examiner to correlate the in situ pathologic features observed in the OCT images to those in the angiographic images. The eye tracking system ensures the precise location of the lesions in both images. The placement of calipers at various points of the Spectralis SD-OCT image allows an analysis and correlation between the 2 imaging modalities, thus confirming the precise location of the lesions, as reported by Coscas and associates. Thus, simultaneous EDI-OCT and fundus fluorescein angiography, or ICG angiography with the Spectralis HRA/OCT system can provide high-quality images with the certainty of knowing the location and exact correlations between the tomographic images of the vascular structures, and the topographic location of the structures can be made in a point-by-point fashion.
The purpose of this study was to determine whether the circle of Zinn-Haller can be seen and studied in highly myopic eyes. Highly myopic eyes were chosen because intrasclerally located blood vessels such as the circle of Zinn-Haller can be observed only in highly myopic eyes with thin retina and choroid by OCT and only in the area of a large conus by ICG angiography. The circle of Zinn-Haller can be displaced from the optic nerve by the mechanical expansion of the peripapillary region of highly myopic eyes. To accomplish this, we performed simultaneous ICG angiography and OCT with the Spectralis HRA/OCT system. The EDI-OCT images showed the intrasclerally situated circle of Zinn-Haller, and the angiographic and anatomic features of the circle of Zinn-Haller were compared on a point-by-point basis.
Materials and Methods
Approval was obtained from the Ethics Committee of the Tokyo Medical and Dental University to perform this retrospective study, and the procedures used during the examinations conformed to the tenets of the Declaration of Helsinki. Written informed consent was obtained from all participants.
Seventy-nine patients (108 eyes) with pathologic myopia underwent ICG angiography and EDI-OCT examinations using Spectralis HRA-OCT (Heidelberg Engineering, Heidelberg, Germany). The patients were examined between February 17, 2012 and March 23, 2012 in the High Myopia Clinic of the Tokyo Medical and Dental University. The definition of pathologic myopia was a refractive error (spherical equivalent) >−8.00 diopters (D) or an axial length >26.5 mm. All of the participants had a comprehensive ocular examination including measurement of the refractive error (spherical equivalent), measurement of the axial length with the IOL Master (Carl Zeiss Meditec, Jena, Germany), and fundus photography (TRC-50DX; Topcon, Tokyo, Japan). The spherical equivalent refractive error was determined as the noncycloplegic values obtained by a table-mounted autorefractor (RK-F1; Canon, Tokyo, Japan).
Indocyanine Green Angiography
To examine the circle of Zinn-Haller by ICG angiography, the early frames (ie, for 2 minutes after the intravenous bolus injection of ICG) were analyzed. The wavelength of the diode laser source was either 790 or 820 nm for the ICG angiography, and the emission wavelengths were limited to wavelengths between 805 and 835 nm. The circle of Zinn-Haller was detected in the video ICG angiographs by one of the authors (M.M.), who was masked to the refractive status and the axial length of the eye.
Enhanced-Depth Imaging Optical Coherence Tomography
Spectral-domain EDI-OCT was performed with the Heidelberg Spectralis HRA+OCT (Heidelberg Engineering) with the acquisition and analysis software (version 5.4) with automated EDI. EDI-OCT was performed using a technique similar to that described in detail by Margolis and Spaide. The axial resolution was 3.5 μm with a scan speed of 40 000 A-scans/second. The scanning protocol consisted of 30 × 25-degree volume scans containing 32 lines of 768 A-scans. Real-time eye tracking by TruTrack Active Eye Tracking (Heidelberg Engineering) was used, and the automatic real-time image averaging was set at 100 images. The EDI-OCT images were obtained through dilated pupils. An internal fixation light was carefully adjusted manually to center the optic disc in the middle of the frame by examining the SLO monitor.
Among the 108 eyes of 79 patients, 6 eyes were excluded because of the poor quality of the EDI-OCT image of the sclera (attributable to dense cataract in 4 of the 6 eyes). In the remaining 102 eyes, 94 eyes of 67 consecutive patients whose circle of Zinn-Haller was observed within the area of the myopic conus by ICG angiography were studied. The demographics of these patients are shown in the Table . A myopic conus was present in all 94 eyes; an annular conus was observed in 53 eyes (56.4%), a temporal conus in 40 eyes (42.6%), and an inferior conus in the remaining eye.
|Age (y), mean ± SD (range)
|56.3 ± 11.2 (32-84)
|Refractive error (D), mean ± SD (range)
|−14.4 ± 4.3 (−8.5 to −23.5)
|Axial length (mm), mean ± SD (range)
|30.7 ± 2.0 (26.6-36.1)
|Posterior staphyloma, eyes (%)
|Type of myopic conus, eyes (%)
ICG angiography showed that the retrobulbar short posterior ciliary arteries were the first to be filled, and they were seen as wide, highly fluorescent, movable vessels that ran through the area of the myopic conus. Immediately following the filling of the short posterior ciliary arteries, the circle of Zinn-Haller surrounding the optic disc was filled ( Figures 1-4 ). Fine centripetal branches from the circle of Zinn-Haller that ran toward the optic nerve were also seen in 20 eyes ( Figure 2 ; Supplemental Figure , available at AJO.com ). The area where the circle of Zinn-Haller was visible was restricted to the area of the myopic conus. In 63 eyes (67%), the circle of Zinn-Haller was circular and surrounded the optic disc (Top middle image of Figure 1 , Top right image of Figures 2 and 3 ). In the remaining 31 eyes (33.0%), the circle of Zinn-Haller had a horizontally long rhomboid shape (Bottom middle image of Figure 1 ) when the circle of Zinn-Haller was visible both nasal and temporal to the optic nerve, or the circle of Zinn-Haller had a horizontally long triangular shape ( Figure 3 ) when only the temporal part of the circle of Zinn-Haller was visible. In these 31 eyes whose circle of Zinn-Haller was not circular, the entry point of the lateral and/or medial short posterior ciliary arteries was at the most distant point from the optic disc margin (Bottom middle image of Figure 1 , and middle image of Figure 3 ).