Morning glory syndrome (MGS) is a congenital disc anomaly that was named by Kindler in 1970 for its resemblance in appearance to the morning glory flower.1 The optic disc is usually enlarged, orange or pink in color with a funnel-shaped excavation and whitish glial tissue covering it, and surrounded by chorioretinal pigmentary disturbance.1
In addition to the disc anomaly, the vascular pattern is often abnormal. The vessels emanate from the edge of the disc as multiple and narrow branches,1 with vascular sheathing, looping, and stretching, as well as increased capillaries around the disc in some cases.2,3 Rojanaporn at el. reported peripheral nonperfusion in four consecutive MGS eyes.4 Additionally, peripheral nonperfusion had been described in 24 eyes with congenital optic nerve anomalies, including MGS, and the secondary complications of fibrovascular proliferation, vitreous hemorrhage, or tractional retinal detachment (RD) were observed in eyes with optic nerve hypoplasia and optic nerve coloboma.5 This report aims to study the peripheral vascular patterns and the secondary complications of peripheral nonperfusion in MGS.
Patients and Methods
Institutional review board approval from Xinhua Hospital was obtained for this retrospective, cross-sectional study. All patients were diagnosed by two pediatric retina specialists (PZ and QZ) at the Department of Ophthalmology, Xinhua Hospital, affiliated to Shanghai Jiaotong University School of Medicine.
A retrospective chart review of patients diagnosed with MGS was conducted between August 1, 2014, and April 30, 2017. This data range was selected on the basis of the availability of widefield angiography in our clinic. As per routine, all patients underwent comprehensive examinations under general anesthesia, including wide-angle retina photography and fluorescein angiography (FA) by the RetCam3 (Natus Medical, Pleasanton, CA) and indirect ophthalmoscope.
Other cavitary optic disc anomalies, such as optic pit, optic disc coloboma, and peripapillary staphyloma, were excluded. Eyes with severe lens opacity or pupil anomalies that blocked the fluorescence of peripheral retinal vascular were excluded.
Data collected from records of patients included sex, gestational age at birth, birth weight, age at angiography, and family history.
In all cases, FA was performed under general anesthesia using a standard pediatric dose (35 mg/10 lb) of intravenous fluorescein sodium (Fluorescite; Alcon, Fort Worth, TX). To obtain a complete view of the peripheral retina, the camera was titled until the ora serrata was partly in view.
When the distance from the vascular terminal to ora serrata was 1.5 disc diameters or greater temporally and 1.0 disc diameters or greater nasally, the nonperfusion was considered abnormal.6 Severity of peripheral nonperfusion was graded based on the location of vascular termination found in FA as follows: (1) mild, if termination was within Zone III (as defined by the International Classification of Retinopathy of Prematurity) and nonperfusion affected only one quadrant; (2) moderate, if nonperfusion area is more extensive than mild but not severe; (3) severe, if termination was within Zone II in all quadrants; (4) extreme, if termination was within Zone I in at least one quadrant.
A total 86 eyes of 74 patients were included. The demographics of this group of patients are described in Table 1. Male/female ratio was 1:1. All the participants were children, with an average age of 35.22 months ± 28.31 months (range: 3 months to 11 years) at angiography. There was no history of prematurity, low birth weight, or familial exudative vitreoretinopathy (FEVR). Twenty-two (29.73%) patients were bilaterally affected, but one eye was enrolled and the fellow eye was excluded in 10 of these patients, among whom eight underwent an operation and another two had severe cataract or pupillary anomaly. Sixteen eyes had RD at angiography. The detachment was limited around the disc or the posterior pole in eight eyes, spread from the disc to the equator in one quadrant in two eyes and to the ora serrata in two quadrants in three eyes, and involved nearly the whole retina in three eyes. Fifteen eyes had exudative RD, and one eye had tractional RD.
Demographics of Patients With MGS
Seventy-three of 86 eyes (84.88%) had peripheral retinal nonperfusion. The avascular retina ranged in size (from a few degrees up to 360°) and distance from the ora serrata. The nonperfusion characteristics, including the severity, the border, and the architectures of vascular termination, are summarized in Table 2. Representative cases are shown in Figures 1 and 2.
Severity of Peripheral Retinal Nonperfusion and Architectures of Vascular Terminations in Eyes With Nonperfusion
(A) Montage fluorescein angiogram (FA) of patient 52 with moderate peripheral nonperfusion and defined vascular terminations. The ora serrata is marked with white arrows. (B) FA of patient 66 with extreme peripheral nonperfusion and leakage from the vascular terminals. (C, D) Clinical images from patient 69. (C) Montage color photograph showing ghost vessel (white arrow) and subretinal proliferation. (D) FA showing poorly defined vascular terminations and mottled transmitted fluorescence from the disc to the peripheral. (E, D) Clinical images from patient 56. (E) Color photograph showing the neovascular, fibrovascular proliferation and tractional retinal detachment. (F) FA showing the leakage on the fibrovascular proliferation.
Fluorescein angiograms showing different vascular terminals. (A) Branch-like terminals. (B) Loops (laser photocoagulation around the disc was performed as a prophylactic treatment to retinal detachment.) (C) Networks. (D) Brush-like terminals. (E) Bulbous capillary terminals (white arrow). (F) Leakage from the vascular-avascular junction.
Mild nonperfusion was the most common degree in this group of patients with MGS, presenting in 36.05% eyes, although 19.77% eyes had moderate nonperfusion and 20.93% had severe nonperfusion. Extreme cases were only presented in seven of 86 eyes (8.14%), four of which had poorly defined terminations (Figure 1D). Among the 17 eyes with moderate nonperfusion, two quadrants were affected in 10 eyes, three quadrants were affected in five eyes, and four quadrants were affected in two eyes. The vascular terminals showed variable architectures, including terminal vascular networks, loops, and branch-like or brush-like terminations (Figures 2A–2D). Among them, branch-like and loop terminations were most commonly seen, presenting in 38.36% eyes with nonperfusion respectively, and brush-like termination was observed in 15.07% eyes, formed by excessively branching vessels. Sixty of 73 eyes (82.19%) with peripheral avascularity showed defined vascular terminations. Thirteen of 73 eyes (17.81%) had an ill-defined margin of vascularity in different degrees with mottled transmitted fluorescence spread from the disc to the peripheral retina (Figure 1D), and with subretinal proliferation in nine eyes, ghost vessels in three eyes, around-disc RD in two eyes, and disc to ora serrata RD in two eyes.
Five eyes (6.85%) in five patients presented with bulbous capillary terminals but no leakage (Figure 2E). Obvious fluorescein leakage was found in six of 73 eyes (8.22%; five patients) at the vascular-avascular junction, whereas the leakage was on the attached retina (Figure 2F) in three eyes (two patients) and on the detached retina (Figure 1F) in three eyes (three patients). Two eyes were found to have fibrovascular proliferation, with tractional detachment in one of them (Figures 1E and 1F).
The nonperfusion area did not appear in 13 of 86 eyes (15.12%), of which, seven eyes had a relatively intact macular area, recognized by the dark appearance in angiograms. Regardless of whether eyes had or did not have peripheral nonperfusion, retinal vessels demonstrated excessive and narrow branching from the enlarged optic disc, radiating to the peripheral area in a straight pattern.
Including the 13 eyes mentioned above, an area with mottled transmitted fluorescence at the posterior pole around the disc in variable size without a demarcated edge was observed in 33 of 86 eyes (38.37%). In these eyes, 32 eyes had peripheral nonperfusion and 15 eyes had RD.
This study provided the largest widefield angiographic survey of a population of patients with MGS to date. In this group, there is no sex predominance, and bilateral patients accounted for nearly 30% of the total, which is higher than other reports.3,7 This may be because unilateral patients were easier to be excluded once with severe complications, whereas bilateral patients were excluded only when both eyes had severe complications.
Peripheral nonperfusion is a significant clinical feature of MGS since it was found in 84.88% eyes in this report. Retinal nonperfusion is the absence of blood flow through all or part of the retina. Secondary nonperfusion, the most common type, is the result of blockage of blood flow, whereas primary nonperfusion is due to the incomplete development of retinal vascularization. It is easy to differentiate between primary and secondary nonperfusion by FA. The vessels in primary nonperfusion show a defined termination, beyond which there is completely avascular area. However, areas of secondary nonperfusion are not avascular and may have ghost vessels. In this group of patients with MGS, nonperfusion with a defined vascular termination in 60 eyes is easy to be defined as primary nonperfusion. Secondary nonperfusion might contribute to the nonperfusion in the other 13 eyes with poorly defined vascular termination, with ghost vessels in three eyes.
Primary nonperfusion is most commonly seen in retinopathy of prematurity (ROP), resulting from environmental disruption of fetal angiogenesis; other causes include FEVR and Norrie disease, both resulting from the genetic mutations that prevent normal angiogenesis in full-term infants. The angiography study of this group of patients with MGS revealed that primary nonperfusion area exists in most patients with MGS, with no evidence of ROP or family history of FEVR and Norrie disease. In embryos, retinal vasculature follows the gradient of retinal ganglion cell maturation and occurs by angiogenesis induced by vascular endothelial growth factor (VEGF), expressed by neuroglia in response to physiologic hypoxia.8 A pathologic study of an eye with MGS revealed that in the retina, the ganglion cell layer was replaced by glial cells and preretinal spindle cells were identified in some areas.9 Since that normal retina maturation is essential for normal retinal vasculature, in eyes with MGS, the maldeveloped retina might not express enough VEGF to complete the retinal vasculature, causing the primary peripheral nonperfusion. The observation that peripheral nonperfusion was absent in seven eyes with relatively intact macular area further proved that eyes with better-developed retina tended to have better-developed retinal vasculature.
Secondary nonperfusion can occur in pediatric eye diseases, including incontinentia pigmenti, Coats' disease, and shaken baby syndrome. In the 13 eyes with secondary nonperfusion in our group, mottled transmitted fluorescence was spread from the disc to peripheral corresponding to the pigment alteration in fundus photographs, which was considered to be caused by the retinal pigment epithelium alteration after chronic RD; moreover, subretinal proliferation was found in nine eyes. Herein, we speculate that RD may lead to secondary nonperfusion in eyes with MGS.
In ROP and FEVR, the avascular retina seems to induce a VEGF response that leads to initial exudation, neovascularization, and ultimately tractional RD. Congenital optic nerve hypoplasia and optic nerve coloboma also have been reported with primary peripheral retinal nonperfusion and the secondary complications of fibrovascular proliferation and tractional RD.5 In this study, retinal leakage at the vascular-avascular junction was found in six eyes with MGS (8.22%; five patients), which never has been reported before, along with bulbous-like vascular changes in five eyes (similar to those found in FEVR), fibrovascular proliferation in two eyes, and tractional RD in one eye. This may illustrate that, even rarely, the nonperfusion may be active in some cases.10
It seems that the nonperfusion is “silent” in most eyes with MGS, as with mild FEVR. There are reports about FEVR-associated rhegmatogenous RD (RRD) causing by the tears or holes, with the fellow eyes of patients with FEVR-RRD, most seeming as asymptomatic or “normal,” characterized by predetachment changes, including lattice degeneration located at the margin of the vascular-avascular junction in FA.11 In our clinic, RRD caused by holes in the lattice degeneration at the vascular-avascular junction in a 12-year-old patient with MGS has been observed, but was not included in this group and not reported yet. There might be some connection between peripheral nonperfusion with lattice degeneration and retinal holes. In this group of patients, the young average age at angiography, 35.22 months, might be the reason for no finding of lattice degeneration. The MGS-associated RRD might be overlooked for the ignorance of peripheral retina.
Major limitations of this study included its retrospective design and follow-up by FA not available. Another limitation is the lack of data about the visual acuity, the depth and width of the excavated disc, and the axial length of the eye, which might be related to the severity of nonperfusion, which needs further research to address.
In conclusion, we highlighted the peripheral nonperfusion and aberrant vascular pattern in most pediatric eyes with MGS, and firstly reported the secondary complications, including leakage, fibrovascular proliferation, and tractional RD in MGS, suggesting that peripheral retina of MGS deserves more concern in the clinic and needs more research.
- Kindler P. Morning glory syndrome: Unusual congenital optic disk anomaly. Am J Ophthalmol. 1970;69(3):376–384. doi:10.1016/0002-9394(70)92269-5 [CrossRef]
- Beyer WB, Quencer RM, Osher RH. Morning glory syndrome. Ophthalmology. 1982;89(12):1362–1367. doi:10.1016/S0161-6420(82)34624-2 [CrossRef]
- Harasymowycz P, Chevrette L, Décarie JC, et al. Morning glory syndrome: Clinical, computerized tomographic, and ultrasonographic findings. J Pediat Ophth Strab. 2005;42(5):290–295.
- Rojanaporn D, Kaliki S, Shields CL, Shields JA. Morning glory disc anomaly with peripheral retinal nonperfusion in 4 consecutive cases. Arch Ophthalmol. 2012;130(10):1327–1330. doi:10.1001/archophthalmol.2012.505 [CrossRef]
- Shapiro MJ, Chow CC, Blair MP, Kiernan DF, Kaufman LM. Peripheral nonperfusion and tractional retinal detachment associated with congenital optic nerve anomalies. Ophthalmology. 2013;120(3):607–615. doi:10.1016/j.ophtha.2012.08.027 [CrossRef]
- Blair MP, Shapiro MJ, Hartnett ME. Fluorescein angiography to estimate normal peripheral retinal nonperfusion in children. J AAPOS. 2012;16(3):234–237. doi:10.1016/j.jaapos.2011.12.157 [CrossRef]
- Fei P, Zhang Q, Li J, Zhao P. Clinical characteristics and treatment of 22 eyes of morning glory syndrome associated with persistent hyperplastic primary vitreous. Br J Ophthalmol. 2013;97(10):1262–1267. doi:10.1136/bjophthalmol-2013-303565 [CrossRef]
- Gariano RF, Gardner TW. Retinal angiogenesis in development and disease. Nature. 2005;438(7070):960–966. doi:10.1038/nature04482 [CrossRef]
- Dempster AG, Lee WR, Forrester JV, McCreath GT. The ‘morning glory syndrome’ — A mesodermal defect?Ophthalmologica. 1983;187(4):222–230. doi:10.1159/000309330 [CrossRef]
- Trese MT, Kashani AH. Advances in the diagnosis, management and pathophysiology of capillary nonperfusion. Expert Rev Ophthalmol. 2012;7(3):281–292. doi:10.1586/eop.12.26 [CrossRef]
- Yuan M, Ding X, Yang Y, et al. Clinical features of affected and undetached fellow eyes in patients with FEVR-associated rhegmatogenous retinal detachment. Retina. 2017;37(3):585–591. doi:10.1097/IAE.0000000000001171 [CrossRef]
Demographics of Patients With MGS
|Mean ± SD|
|Age at Angiography||35.22 months ± 28.31 months|
Severity of Peripheral Retinal Nonperfusion and Architectures of Vascular Terminations in Eyes With Nonperfusion
|Severity of Nonperfusion||N = 86|
|Architectures of Vascular Termination in Eyes With Nonperfusion||N = 73|
|Poorly defined border||13||17.81|
|Bulbous capillary terminals||5||6.85|