You are here

Article Text

Article menu

Download PDFPDF

Clinical science
Choroidal filling delay in choroidal neovascularisation due to pathological myopia
  1. Taku Wakabayashi,
  2. Yasushi Ikuno
  1. Department of Ophthalmology, Osaka University Medical School, Osaka, Japan
  1. Correspondence to Dr Yasushi Ikuno, Department of Ophthalmology, Osaka University Medical School, Room E7, 2-2 Yamadaoka, Suita 565-0871, Osaka, Japan; ikuno{at}ophthal.med.osaka-u.ac.jp

Abstract

Aim To assess the choroidal thickness and choroidal circulatory changes in eyes with myopic choroidal neovascularisation (mCNV).

Methods Retrospective, consecutive, observational case series. Forty-two consecutive eyes (17 eyes with newly diagnosed mCNV and 25 eyes without CNV) were included. Choroidal circulation was evaluated by indocyanine green angiography (ICGA), and choroidal thickness was measured by spectral-domain optical coherence tomography (SD-OCT). The factors associated with mCNV were evaluated.

Results Sixteen (94%) of 17 eyes with mCNV and six (24%) of 25 eyes without mCNV had well-defined hypofluorescence at the macular region on arterial phase ICGA, that is, a choroidal filling delay. Older age (p<0.001), the presence of a choroidal filling delay (p<0.001) and reduced choroidal thickness (p=0.003) were significantly associated with mCNV on univariate analysis. The most important of these three factors associated with mCNV, in order of importance, were the choroidal filling delay (OR=41.5, p<0.001) and choroidal thickness (per 1 μm, OR=0.97, p=0.01). Older age was significantly associated with both choroidal filling delay (per 1 year, OR=1.16, p<0.001) and choroidal thinning (regression coefficient=−1.22, p<0.001).

Conclusions Significant choroidal changes were observed in eyes with mCNV. Ischaemia-induced growth factor expression caused by decreased choroidal perfusion may be related to the development of mCNV.

  • Myopic choroidal neovascularisation
  • choroidal filling delay
  • choroidal thinning
  • indocyanine green angiography
  • choroid
  • retina
  • neovascularisation
  • imaging

https://doi.org/10.1136/bjo.2009.163535

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Choroidal neovascularisation (CNV) is a major cause of severe vision loss in patients with pathological myopia. Myopic CNV occurs in 4–11% of patients with high myopia,1 with most eyes progressing to 20/200 or worse within 5–10 years after onset.2 3 The pathogenesis of myopic CNV is not known. A close relationship with lacquer crack (LC) formation has been reported.4 5 LCs are ubiquitously observed in highly myopic eyes, but the specificity of LCs for mCNV is unclear. Photodynamic therapy6 7 and intravitreal bevacizumab (Avastin, Genentech, San Francisco, California)8 9 are the current major treatments for myopic CNV; however, the long-term treatment effects are not favourable due to progressive chorioretinal macular atrophy in the late phase. Therefore, investigation of the pathogenesis of mCNV is crucial to establish strategies for prevention.

    Choroidal circulatory disturbances occur in CNV or CNV-related cases such as age-related macular degeneration (AMD), polypoidal choroidal vasculopathy (PCV) and central serous chorioretinopathy (CSC).10–12 For example, a choroidal filling delay, recognised as a well-defined hypofluorescent region, usually of irregular shape, is observed in 32% of early AMD cases by indocyanine green angiography (ICGA).10 These findings indicate that choroidal vascular abnormalities may correlate significantly with the occurrence of CNV.

    Laser Doppler velocimetry (LDV) and colour Doppler ultrasonography studies in humans and animals suggest that choroidal blood flow is disturbed in highly myopic eyes.13 14 In addition, several histopathological and angiographic studies have demonstrated pronounced thinning and disappearance of the lamina of small choroidal vessels in highly myopic eyes.15–19 Choroidal circulatory status, however, has not been investigated to any great degree in highly myopic eyes with a specific focus on mCNV.

    ICGA combined with confocal scanning laser ophthalmoscope using the Heidelberg Retina Angiograph 2 (HRA2, Heidelberg Engineering, Heidelberg, Germany) provides higher-contrast images than a conventional ICG fundus camera and allows for enhanced imaging of the choroidal vessels. The high-speed video documents the early filling stages and may enable detection of faint circulatory changes in vivo. This study was designed to investigate the choroidal circulatory changes and its risk factors for mCNV.

    Patients and methods

    Patients

    This study was an observational case series of 42 eyes of 42 consecutive patients who visited the Myopia Clinic of Osaka University Hospital, Osaka University Medical School, Osaka Japan, and underwent HRA2-ICGA for macular diseases associated with pathological myopia. Pathological myopia was defined as eyes with a spherical equivalent refractive error greater than −6.0 dioptres or axial length more than 26.5 mm. Exclusion criteria included features suggesting that the CNV may be associated with AMD, angioid streaks or multifocal choroiditis. Eyes with severe chorioretinal atrophy with transmission of sclera were also excluded. The presence or absence of myopic CNV was determined using FA and the consensus of the two authors (TW and YI). All patients underwent comprehensive ophthalmological examinations, including best-corrected visual acuity (BCVA), slit-lamp biomicroscopy with a contact lens, fundus photography, FA, ICGA and optical coherence tomography (OCT). ICGA was performed with a confocal scanning laser ophthalmoscope, the HRA2, by two masked examiners. Written informed consent was obtained from all patients. Approval from the institutional review board was not required for this observational study.

    Indocyanine green angiography

    In ICGA, patients received a 25 mg bolus of ICG. An infrared diode laser (795 nm) was used for excitation, and emission was detected with a barrier filter at 810 nm. The scanning field was a 30° rectangle. The sequence of early filling of the choroidal vessels was recorded using the movie mode during the first minute after ICG infusion. Subsequent single images were obtained for late ICGA images at 5, 10 and 20 min. The filling delay was defined as a distinct area of prolonged circulatory defect involving the fovea at 15–25 s, that is, the arterial and arteriovenous phase that was gradually filled in a venous phase. The presence/absence of choroidal filling delay was determined by TW and YI. An interobserver agreement was achieved in 41 eyes (98%). Those areas were differentiated by chorioretinal atrophy presenting hypofluorescence, traversed only by large choroidal vessels, through the venous phase in ICGA.

    Optical coherence tomography

    OCT images were obtained by high-definition (HD) spectral-domain OCT, the Cirrus HD-OCT (Carl Zeiss Meditec, Dublin, California), in all cases. The scan patterns included the five-line raster mode and the macular cube 512×128 combo mode. The five-line raster provides five closely spaced horizontal lines. Each line has a 6 mm range and comprises 4096 A-scans. The macular cube 512×128 combo mode generates a cube of data through a 6 mm square grid by acquiring a series of 128 horizontal scan lines comprising 512 A-scans. Because the relatively thin retinal and choroidal thickness in highly myopic eyes allows deep penetration of the light beneath the retinal pigment epithelium and visualisation of sclera/choroid interface, choroidal thickness was measured as the distance between the highly reflective layer representing the retinal pigment epithelium (RPE) and the reflective line of the sclera/choroid interface beneath the fovea.20 The measurement of the choroidal thickness was conducted manually by two masked observers using the software included in the Cirrus HD-OCT, as previously described,20 and the values of the measurements were averaged.

    Statistical analysis

    The t test, Mann–Whitney rank sum test and Fisher exact test were conducted when appropriate to assess the univariate association of risk factors with mCNV. Risk factors with a p value ≤0.25 were considered to be candidate risk factors for the stepwise regression analysis. Logistic regression analyses were conducted to identify the OR for factors identified as significant in the stepwise regression analysis. All analyses were performed using the JMP statistical software package (V.7.0, SAS Institute, Cary, North Carolina). A p value of less than 0.05 was considered statistically significant.

    Results

    Of the 42 patients (15 men, 27 women), 17 eyes had newly diagnosed mCNV, and 25 eyes did not have mCNV (18 eyes with simple subretinal haemorrhage (SH) and seven eyes with neither SH nor CNV). The mean age was 46.2±16.1 years (range 17–73). Typical myopic changes were observed in all eyes on colour fundus photographs.

    Twenty-two (52%) eyes had a well-defined hypofluorescent area at the macular area at 15–25 s (arterial and arteriovenous phase) in HRA2-ICGA, that is, a choroidal filling delay (figures 1, 2). These areas became less prominent at 30 s and were filled further with ICG within 1 min, except for areas with LC formation. In the other 20 (48%) eyes, no choroidal filling delay at the macular area was observed, although in four eyes there was a similar filling delay at the extramacular area, especially the peripapillary region. The LC formation was observed in 39 (93%) of 42 study eyes (16 (94%) of 17 eyes with mCNV and 23 (92%) of 25 eyes without mCNV). Of the seven eyes with neither SH nor CNV, six eyes had LC formation and one eye had no abnormal findings on HRA2-ICGA. Also, none of these seven eyes presented typical round/oval chorioretinal myopic atrophy or fibrosis. The Cirrus HD-OCT clearly demonstrated the reflective line of the sclera/choroid interface in all study eyes. The mean choroidal thickness was 80.5±34.3 μm (range 22–171 μm).

    Figure 1
    Figure 1

    Colour fundus photograph (A), mid-phase fluorescein angiogram (FA) (B), spectral-domain optical coherence tomography (SD-OCT) (C), artery-phase (D), arteriovenous phase (E) and venous-phase (F) indocyanine green angiography using the Heidelberg Retina Angiogram 2 (HRA2-ICGA) of a 53-year-old man with a choroidal filling delay along with myopic choroidal neovascularisation (mCNV). (A) Fundus photograph showing a typical myopic fundus with a temporal crescent at the optic disc and a visible choroidal vascular pattern. A small greyish subfoveal lesion is presumed to be mCNV. (B) Mid-phase FA showing a small hyperfluorescent area corresponding to the CNV. There is no apparent window defect at the fovea. (C) SD-OCT image showing juxtafoveal CNV. The hyper-reflective lines represent the retinal pigment epithelium (yellow arrowheads) and chorio-scleral interface (white arrowheads), and the distance between the two lines, indicating choroidal thickness, was measured beneath the fovea. (D) HRA2-ICGA clearly showing a filling delay in the artery-phase (arrowheads). (E) Arteriovenous phase. The area is perfused and a filling delay is no longer prominent. (F) Venous phase. The filling delay is no longer apparent. Small horizontal lacquer cracks (arrow) are present.

    Figure 2
    Figure 2

    Colour fundus photograph (A), mid-phase fluorescein angiogram (FA) (B), spectral-domain optical coherence tomography (SD-OCT) (C), artery-phase (D), arteriovenous phase (E) and venous-phase (F) indocyanine green angiography using the Heidelberg Retina Angiogram 2 (HRA2-ICGA) of a 59-year old female with a choroidal filling delay along with myopic choroidal neovascularisation (mCNV). (A) Fundus photograph showing a typical myopic fundus with a visible choroidal vascular pattern. A small greyish subfoveal lesion is presumed to be mCNV. (B) Mid-phase FA showing a small hyperfluorescent area corresponding to the CNV. There is no apparent window defect at the fovea. (C) SD-OCT image showing juxtafoveal CNV. The hyper-reflective lines represent the retinal pigment epithelium (yellow arrowheads) and chorio-scleral interface (white arrowheads), and the distance between the two lines, indicating choroidal thickness, was measured beneath the fovea. (D) HRA2-ICGA clearly showing a filling delay in the artery phase (arrowheads). (E) Arteriovenous phase. The area is perfused and a filling delay is no longer prominent. (F) Venous phase. The filling delay is no longer apparent.

    Table 1 shows the univariate associations of various factors with mCNV. Older age (p<0.001), the presence of a choroidal filling delay (p<0.001) and thinner choroidal thickness (p=0.003) were all significantly associated with mCNV. Candidate risk factors with a p value ≤0.25, including age, axial length, choroidal filling delay and choroidal thickness, were entered into the stepwise regression analysis to assess the most significant factors for mCNV. The factors associated with mCNV, in order of importance, were a choroidal filling delay and choroidal thickness (R2=0.57, p<0.01, p <0.05, respectively). A logistic regression analysis was then performed. A choroidal filling delay was associated with a 41.5-fold higher risk for mCNV (p<0.001) (table 2). In addition, there was a 0.97-fold risk for CNV associated with every 1 μm increase in choroidal thickness (p=0.01).

    View this table:
    Table 1

    Univariate associations with myopic choroidal neovascularisation (mCNV)

    View this table:
    Table 2

    Factors associated with the presence of myopic choroidal neovascularisation based on stepwise regression analysis and logistic regression analysis

    We further analysed the factors associated with a choroidal filling delay and choroidal thickness. When assessing the risk of a choroidal filling delay using a stepwise regression analysis, age was the factor associated with a choroidal filling delay (R2=0.44, p<0.001), but not choroidal thickness, axial length or sex. The logistic regression analysis identified a 1.16-fold risk for a choroidal filling delay in association with every 1-year increase in age (p<0.001). Stepwise regression analysis also identified age as a factor associated with choroidal thinning (R2=0.43, p <0.001), but not a choroidal filling delay, axial length or sex. Multiple linear regression analysis revealed a correlation between choroidal thinning and age with a regression coefficient of −1.22 (p<0.001).

    Discussion

    Numerous studies have described the postmortem histology of highly myopic eyes. Most studies report the following findings: (1) disappearance of large vessels and/or capillaries, (2) displacement by fibrous tissue, (3) disappearance or disturbance of the RPE layer and (4) consequent loss of the photoreceptors and outer retinal layers after severe choroidal atrophy.15–18 However, histological investigation in postmortem eyes is limited. Therefore, the HRA2-ICGA system is a powerful modality for studying the choroidal circulation status in vivo. Non-invasive examinations such as laser Doppler velocimetry (LDV) or colour Doppler ultrasonography suggest that highly myopic eyes have decreased choroidal blood flow.13 14 The scan area is limited in LDV studies. Laser speckle flowmetry is another option, but in general, the data cannot be compared among individuals. With ICGA, visualisation of an extended area of the posterior eye portion is possible, thus allowing for better detection of choroidal circulatory abnormalities in a broader region.

    Univariate analysis indicated that older age, presence of a choroidal filling delay and choroidal thinning were significantly associated with mCNV. A stepwise regression analysis found that a choroidal filling delay was the most significant factor, and choroidal thinning was the second strongest factor associated with mCNV, but age was not a factor. These findings suggest that choroidal changes are more significantly associated with mCNV than is age. These three factors are confounding, however, as described below. To rule out the influence of age, a case-control study must be considered, but because HRA-ICGA is an invasive examination, it should not be applied for the examination of healthy myopic eyes for ethical reasons.

    Stepwise analysis indicated that a choroidal filling delay was the factor most strongly associated with mCNV. A logistic regression model revealed a 41.5-fold higher risk for mCNV in the presence of a filling delay. Choroidal filling delays are observed in other CNV-associated disorders,10–12 and a similar mechanism may be involved in the development of mCNV. The mechanism is unclear; however, the choroidal circulatory disturbance may induce hypoxia in the outer retinal part including photoreceptors, Muller cells and RPE, which is a large source of vascular endothelial growth factor (VEGF) secretion. Because anti-VEGF agents are effective for treating myopic CNV,8 9 mCNV may be somewhat dependent on VEGF.

    Choroidal thickness was also significantly associated with mCNV by univariate analysis, and the second strongest factor in the stepwise regression analysis. Histopathological studies indicate that choroidal thinning may be due to choroidal vascular disappearance and thus may be related to choroidal vascular occlusion. In our analysis, however, choroidal thinning and a filling delay were not strongly correlated, and so we believe that the choroidal filling delay observed in the present study was not due simply to choroidal vascular loss with thinning but rather to the coexistence of an abnormal vascular shape or flow speed changes. We were unable to quantitate the choroidal blood flow by ICG, but this would provide additional information about how a filling delay occurs.

    In the present study, univariate analysis indicated that age was a significant factor for developing mCNV. The stepwise regression analysis, however, did not select age as a significant factor for mCNV, but instead selected as a significant factor for both a choroidal filling delay and choroidal thickness. Based on histological examination, choroidal thickness is negatively correlated with age.21 These findings lead us to hypothesise that ageing may cause a choroidal filling delay and choroidal thinning, and that these choroidal changes subsequently induce the development of mCNV (figure 3). Age-induced decreases in choroidal blood flow, choroidal thickness and number of choroidal arterioles have also been reported in other clinical and histological studies.20–22

    Figure 3
    Figure 3

    Hypothesis of the association between age, a choroidal filling delay, choroidal thinning, vascular endothelial growth factor (VEGF) and myopic choroidal neovascularisation (mCNV).

    Interestingly, choroidal filling delay was unique, in terms of its location. The filling delay was most prominent in nasal and around the macula. There was a variation in its shape among patients, but the shape was vertical temporal to the optic nerve disc and horizontal at the macula, which is similar to choroidal watershed zone, proposed by Hayrah.23 24 The watershed zone of the choroid is typically recognised as a hypofilling area of the choroid in angiography.23 The watershed zone indicates the isolation of a choroidal capillary bed supplied by an independent posterior cilliary artery (PCA) that does not anastomose with another end-artery, making it prone to ischaemia. One to five PCA branches arise from the ophthalmic artery.24 There are many interindividual variations, but the posterior choroid is supplied by two major PCAs, the lateral and medial PCA, in approximately 90% of eyes.24 A watershed zone between the medial and lateral PCA is observed vertically between the macula and optic nerve head in 60% of eyes.23 Also, the lateral watershed zone is sometimes observed at the macula in case of multiple lateral PCAs, but two or more lateral PCAs are observed only in 22% of the human subjects24 during the choroidal filling delay in 94% in mCNV cases in this study. Also, choroidal filling delay was observed in an irregular shape, and this is unlike the watershed zone recognised as a relatively linear straight lesion.23 24 Therefore, we speculate that choroidal filling delay might be originated from the watershed zone at least in part, but they are not totally identical.

    The limitations of this study included its retrospective design and a limited number of study eyes. Several clinical studies have reported a close relationship between LC formation and the development of mCNV. The incidence of LC is shown to be higher in eyes with CNV than those without CNV, reaching 75–94% of eyes with CNV.4 5 It is also known that posterior staphyloma formation is also associated with mCNV.25 It is still uncertain if choroidal filling delay is associated with these factors. In addition, because HRA2-ICGA is invasive, we could not compare the patients with healthy volunteers, which made it difficult to evaluate the specificity of a choroidal filling delay and choroidal thinning for mCNV. Nevertheless, the present study revealed the presence of choroidal changes and their potential contribution to mCNV. The development of novel therapeutic strategies to maintain the normal choroid may contribute to the prevention of mCNV.

    Acknowledgments

    The authors acknowledge the immense contribution of the late Y Tano to the present work.

    References

    1. ↵
      1. Ryan SJ
      1. Soubrane G,
      2. Coscas GJ
      . Choroidal neovascular membrane in degenerative myopia. In: Ryan SJ, ed. Retina. 4th edn. St Louis: Mosby, 2005:1136–52.
    2. ↵
      1. Avila MP,
      2. Weiter JJ,
      3. Jalkh AE,
      4. et al
      . Natural history of choroidal neovascularization in degenerative myopia. Ophthalmology 1984;91:1573–81.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Yoshida T,
      2. Ohno-Matsui K,
      3. Yasuzumi K,
      4. et al
      . Myopic choroidal neovascularization: a 10-year follow-up. Ophthalmology 2003;110:1297–305.
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Ohno-Matsui K,
      2. Yoshida T,
      3. Futagami S,
      4. et al
      . Patchy atrophy and lacquer cracks predispose to the development of choroidal neovascularisation in pathological myopia. Br J Ophthalmol 2003;87:570–3.
      OpenUrlAbstract/FREE Full Text
    5. ↵
      1. Ikuno Y,
      2. Sayanagi K,
      3. Soga K,
      4. et al
      . Lacquer crack formation and choroidal neovascularization in pathological myopia. Retina 2008;28:1124–31.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      VIP Study Group. Verteporfin therapy of subfoveal choroidal neovascularization in pathologic myopia: 2-year results of a randomized clinical trial–VIP report no. 3. Ophthalmology 2003;110:667–73.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Krebs I,
      2. Binder S,
      3. Stolba U,
      4. et al
      . Choroidal neovascularization in pathologic myopia: three-year results after photodynamic therapy. Am J Ophthalmol 2005;140:416–25.
      OpenUrlPubMed
    8. ↵
      1. Sakaguchi H,
      2. Ikuno Y,
      3. Gomi F,
      4. et al
      . Intravitreal injection of bevacizumab for choroidal neovascularisation associated with pathological myopia. Br J Ophthalmol 2007;91:161–5.
      OpenUrlAbstract/FREE Full Text
    9. ↵
      1. Chan WM,
      2. Lai TY,
      3. Liu DT,
      4. et al
      . Intravitreal bevacizumab (Avastin) for myopic choroidal neovascularization: six-month results of a prospective pilot study. Ophthalmology 2007;114:2190–6.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Pauleikhoff D,
      2. Spital G,
      3. Radermacher M,
      4. et al
      . A fluorescein and indocyanine green angiographic study of choriocapillaris in age-related macular disease. Arch Ophthalmol 1999;117:1353–8.
      OpenUrlPubMedWeb of Science
    11. ↵
      1. Sasahara M,
      2. Tsujikawa A,
      3. Musashi K,
      4. et al
      . Polypoidal choroidal vasculopathy with choroidal vascular hyperpermeability. Am J Ophthalmol 2006;142:601–7.
      OpenUrlCrossRefPubMed
    12. ↵
      1. Iida T,
      2. Kishi S,
      3. Hagimura N,
      4. et al
      . Persistent and bilateral choroidal vascular abnormalities in central serous chorioretinopathy. Retina 1999;19:508–12.
      OpenUrlCrossRefPubMed
    13. ↵
      1. Shih YF,
      2. Fitzgerald ME,
      3. Norton TT,
      4. et al
      . Reduction in choroidal blood flow occurs in chicks wearing goggles that induce eye growth toward myopia. Curr Eye Res 1993;12:219–27.
      OpenUrlPubMedWeb of Science
    14. ↵
      1. Akyol N,
      2. Kükner AS,
      3. Ozdemir T,
      4. et al
      . Choroidal and retinal blood flow changes in degenerative myopia. Can J Ophthalmol 1996;31:113–19.
      OpenUrlPubMedWeb of Science
    15. ↵
      1. Duke-Elder S
      1. Duke-Elder S
      . Pathological refractive errors: Ophthalmic optics and refraction. In: Duke-Elder S, ed. System of Ophthalmology. St Louis: Mosby, 1970:297–373.
    16. ↵
      1. Ohno H
      . Electron microscopic studies of myopic retinochoroidal atrophies 1. Choroidal changes. Folia Ophthalmologica Japonica (in Japanese) 1983;43:1244–53.
      OpenUrl
    17. ↵
      1. Matsuo N,
      2. Okabe S,
      3. Hasegawa E,
      4. et al
      . Report of research committee on chorioretinal degenerations, The Ministry of Health and Welfare of Japan (in Japanese). 1978.
    18. ↵
      1. Tokoro T
      1. Tokoro T
      . Diffuse chorioretinal atrophy. In: Tokoro T, ed. Atlas of posterior fundus changes in pathological myopia. Tokyo: Springer, 1998:7–14.
    19. ↵
      1. Quaranta M,
      2. Arnold J,
      3. Coscas G,
      4. et al
      . Indocyanine green angiographic features of pathologic myopia. Am J Ophthalmol 1996;122:663–71.
      OpenUrlPubMedWeb of Science
    20. ↵
      1. Ikuno Y,
      2. Tano Y
      . Retinal and choroidal biometry in highly myopic eyes using spectral-domain optical coherence tomography. Invest Ophthalmol Vis Sci 2009;50:3876–80.
      OpenUrlAbstract/FREE Full Text
    21. ↵
      1. Ramrattan RS,
      2. van der Schaft TL,
      3. Mooy CM,
      4. et al
      . Morphometric analysis of Bruch's membrane, the choriocapillaris, and the choroid in aging. Invest Ophthalmol Vis Sci 1994;35:2857–64.
      OpenUrlAbstract/FREE Full Text
    22. ↵
      1. Ito YN,
      2. Mori K,
      3. Young-Duvall J,
      4. et al
      . Aging changes of the choroidal dye filling pattern in indocyanine green angiography of normal subjects. Retina 2001;21:237–42.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. Hayreh SS
      . In vivo circulation and its watershed zones. Eye 1990;4:273–89.
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Hayreh SS
      . The ophthalmic artery III branches. Br J Ophthalmol 1962;46:212–47.
      OpenUrlFREE Full Text
    25. ↵
      1. Steidl SM,
      2. Pruett RC
      . Macular complications associated with posterior staphyloma. Am J Ophthalmol 1997;123:181–7.
      OpenUrlPubMedWeb of Science

    Footnotes

    • Competing interests None.

    • Patient consent Obtained.

    • Provenance and peer review Not commissioned; externally peer reviewed.