Vitreous Changes in High Myopia Observed by Swept-Source Optical Coherence Tomography

Abstract

Purpose.: To observe vitreous changes in high myopia using swept-source optical coherence tomography (SS-OCT).

Methods.: We performed slit-lamp biomicroscopy and SS-OCT in the highly myopic right eyes of 151 patients (mean age, 52.7 years; mean refraction, −11.4 diopters [D]) and the right eyes with no myopia of 363 healthy control volunteers (mean age, 52.8 years; mean refraction, −1.4 D). To estimate the sizes of the posterior precortical vitreous pockets (PPVPs), we measured the height between the fovea and the anterior border of the PPVPs.

Results.: Patients with partial posterior vitreous detachments (PVDs) around the macula and complete PVDs in high myopia were significantly (P < 0.0001) younger (47.1 ± 14.1 and 61.2 ± 12.0 years, respectively) than controls (59.0 ± 9.6 and 69.7 ± 6.6 years). The PPVPs with no PVDs were significantly (P < 0.001) higher in 32 eyes with high myopia (984 ± 292 μm) than 164 controls (553 ± 166 μm). After a complete PVD with a Weiss ring developed, the vitreous cortex was on the macula in 40.5% of the eyes with high myopia, which differed significantly (P < 0.0001) from the 8.7% of the controls. Myopic foveoschisis was present in 14 (9.3%) of 151 eyes. In eyes with foveoschisis, three (21.4%) eyes had partial PVDs and 11 (78.6%) eyes had complete PVDs; there was no residual cortex in 8 (72.7%) of 11 eyes with complete PVDs.

Conclusions.: Highly myopic eyes may have larger PPVPs than normal eyes. Partial PVDs around the macula and complete PVDs occur at younger ages. The vitreous cortex more frequently remains on the macula after development of complete PVDs in highly myopic eyes.

Introduction

High myopia is associated with vitreous liquefaction, and posterior vitreous detachment (PVD) occurs at a younger age than in nonmyopic eyes. ^1–3 The incidence of rhegmatogenous retinal detachment is higher in patients with axial myopia. ⁴ Optical coherence tomography (OCT) showed that foveal retinoschisis frequently develops in highly myopic eyes with posterior staphyloma. ⁵ Because of the transparent tissue and advanced liquefaction, biomicroscopy of the vitreous is difficult, particularly in highly myopic eyes. There are frequent discrepancies between the biomicroscopic and intraoperative findings of the vitreous in highly myopic eyes. Vitreous surgeons occasionally encounter residual cortex on the retina during the surgery performed to treat myopic foveoschisis despite the apparent PVD with a Weiss ring. The anatomy of the vitreous has been studied using biomicroscopy in postmortem eyes. We previously reported the presence of a posterior precortical vitreous pocket (PPVP), which is a boat-shaped vitreous lacuna in front of the posterior pole, at autopsy in eyes in which the vitreous gel was stained with fluorescein. ⁶ Although the PPVP has a key role in the development of various vitreomacular disorders, ^7,8 providing direct proof was difficult because the PPVPs were not visible on slit-lamp biomicroscopy. The PPVP was recently visualized during vitrectomy that included intravitreal injection of triamcinolone acetonide. ^9,10 Spectral-domain OCT (SD-OCT) with noise reduction first allowed in vivo observation of the PPVP. ^11–13 Using SD-OCT, we reported that the presence of partial PVD with the PPVP around the macula is a precursor of complete PVD. ¹⁴ It was reported that a perifoveal vitreous detachment is the primary pathogenic event in idiopathic macular hole formation. ^15,16  

Recently introduced swept-source OCT (SS-OCT) clearly showed the PPVP increasing along with increasing the myopic refractive error. ¹⁷  

The object of the current study was to identify the vitreous changes in highly myopic eyes using SS-OCT. 

Methods

We performed SS-OCT (DRI OCT-1 Atlantis; Topcon, Tokyo, Japan) using 12-mm horizontal and vertical scans through the fovea in the right eyes of 151 patients (61 men, 90 women) of varying ages with high myopia over −8.0 diopters (D). The subjects' ages ranged from 20 to 79 years (average, 52.7 ± 16.0 years; 17 eyes, 20–29 years; 19 eyes, 30–39 years; 28 eyes, 40–49 years; 23 eyes, 50–59 years; 40 eyes, 60–69 years; and 24 eyes, 70–79 years). The refractive powers, measured using the RT-6000 Auto Ref-Topography (Tomey, Nagoya, Japan), ranged from −26.0 to −8.0 D (average, −11.4 ± 3.3 D). All eyes were phakic. Cases with a retinal detachment were excluded because the vitreous was highly liquefied. 

We also performed SS-OCT in the right eyes of 363 healthy control volunteers (195 men, 168 women) without high myopia and other vitreoretinal diseases. The subjects' ages ranged from 21 to 79 years (average, 52.8 ± 17.6 years; 60 eyes, 20–29 years; 36 eyes, 30–39 years; 48 eyes, 40–49 years; 56 eyes, 50–59 years; 93 eyes, 60–69 years; and 70 eyes, 70–79 years). The refractive powers ranged from −7.75 to +3.5 D (average, −1.4 ± 2.4 D). All eyes were phakic. The Table shows the profiles of the subjects and controls. 

Table.

 

Characteristics of the Patients With High Myopia and Control Subjects

	Patients With High Myopia, n = 151	Controls With Normal Eyes, n = 363
Sex
Men	61 eyes of 61 cases	195 eyes of 195 cases
Women	90 eyes of 90 cases	168 eyes of 168 cases
Age	20–79 y old, average, 52.7 ± 16.0 y	21–79 y old, average, 52.8 ± 17.6 y old
Refraction, D	−26.0 to −8.0, average −11.4 ± 3.3	−7.75 to +3.5, average −1.4 ± 2.4

This study included consecutive subjects examined from August 2012 to July 2013. Before OCT was performed, all subjects were evaluated for the presence of a PVD using biomicroscopy with a super-field lens (Volk Optical, Inc., Mentor, OH). We defined a complete PVD as a detached posterior vitreous cortex with a Weiss ring. The status of the posterior wall of the PPVP was classified into one of three stages: no, partial, or complete PVD according to the biomicroscopic findings and SS-OCT images. Partial PVD was defined as a paramacular PVD, a perifoveal PVD, or vitreofoveal separation as described previously ¹⁴ (Fig. 1). According to the region of PVD, we distinguished paramacular PVD from perifoveal PVD. Paramacular PVD is outside of the macula, and perifoveal PVD is within the macula. We defined the vitreous cortex remaining on the retina after occurrence of complete PVD with Weiss ring as the residual vitreous cortex. 

Figure 1

 

OCT findings of partial PVD. The status of the posterior wall (arrows) of the posterior precortical vitreous pockets is classified into one of three stages: no PVD, partial PVD, or complete PVD according to the biomicroscopic findings and OCT images. Partial PVD consists of a paramacular PVD (top), a perifoveal PVD (middle), and vitreofoveal separation with persistent attachment to the optic disc (bottom). Paramacular PVD is outside of the macula and perifoveal PVD is within the macula. P denotes a posterior precortical vitreous pocket.

We performed the OCT for all cases with the individual in a seated position. During the OCT examination, we placed the scanner head backwards to obtain the vitreous structure in the B-scan images. The retina and the choroid were positioned at the lower end of the screen to increase the visibility of the vitreous. After obtaining the OCT images, we adjusted the contrast to visualize the gel, liquefied pocket, and the cortex. To estimate the sizes of the PPVP with no PVD, we measured the height between the fovea and the anterior border of the pocket in cases with no PVD, according to our previous report. ¹⁷ Because it is difficult to measure the sizes of deformed pockets with partial and complete PVD, we measured cases with no PVD. If the anterior border was not identified within the scope of the OCT image, we measured the length between the fovea and the edge of the scope. 

A test of normal distribution was performed to decide whether to use Student's t-test or Mann-Whitney U test. The difference of ages in four groups (all cases, no PVD, partial PVD, and complete PVD group) between patients with high myopia and controls was tested by the unpaired, two-tailed Student's t-test. Mann-Whitney U test was used to determine statistical significance of PPVP height between eyes with high myopia and controls in the no PVD group. Chi-square tests were used for the statistical analyses of the incidences of residual vitreous cortex and foveoschisis between myopia and controls. Data are plotted as mean ± standard deviation, and P < 0.05 was considered significant. 

This study was performed under approval of Gunma University Hospital Institutional Review Board. Informed consent for participation in this research was obtained from all patients. All individuals provided informed consent after having received a detailed explanation of the purpose of the study. 

Results

The status of the posterior wall of the PPVPs in 151 highly myopic eyes was classified as no PVD in 32 eyes, partial PVD in 35 eyes, and complete PVD in 84 eyes. In 363 healthy control eyes, the status was classified as no PVD in 164 eyes, partial PVD in 95 eyes, and complete PVD in 104 eyes. Figure 2 shows the percentages of each stage of PVD in each decade of life. 

Figure 2

 

The percentages of each stage of PVD in each decade of life. One hundred fifty-one eyes with high myopia (top). Three hundred sixty-three eyes of healthy control volunteers (bottom).

Figure 3 shows the comparisons of age between patients with high myopia and controls in each of four groups (all cases, no PVD, partial PVD, and complete PVD). In the all cases and no PVD groups, there were no significant differences (Student's t-test: P > 0.05) for age differences between patients with high myopia (52.7 ± 16.0 years in all 151 patients; 36.4 ± 11.6 years in 32 cases of no PVD) and controls (52.8 ± 17.6 years in all 363 controls, 38.5 ± 13.9 years in 84 controls of no PVD). Thirty-five patients with high myopia with a partial PVD (47.1 ± 14.1 years) were significantly (P < 0.001) younger than 95 control subjects with a partial PVD (59.0 ± 9.6 years). Eighty-four patients with high myopia with a complete PVD (61.2 ± 12.0 years) were significantly (P < 0.001) younger than 104 control subjects with a complete PVD (69.7 ± 6.6 years). 

Figure 3

 

Mean ages of the patients with high myopias and controls. In the groups of all cases and no PVD, their ages did not show different between high myopias and controls (P = 0.95; P = 0.44). In the groups of partial and complete PVD, their ages showed strongly different (P < 0.001; P < 0.001). (Student's t-test. P < 0.05 was considered statistically significant. *P < 0.05, **P < 0.01, ***P < 0.001; ns, no statistical significance.)

The PPVPs in 32 eyes with high myopia with no PVD were significantly (P < 0.001) higher (984 ± 292 μm) than in the controls with no PVD (553 ± 166 μm) (Fig. 4). In eight of 32 eyes, we tentatively defined the height of the PPVP as the length between the fovea and the superior edge of the OCT image, because the PPVPs were extremely large and the anterior border could not be seen in the OCT images. 

Figure 4

 

Swept-source OCT images showing the PPVPs in the highly myopic eye and the control. A large PPVP in the right eye of a 33-year-old man with −9.0 D of refractive error (top). A normal PPVP and Cloquet canal in the right eye of a 30-year-old man with −3.5 D of refractive error (bottom). A connecting channel (arrow) is seen in the septum between the Cloquet canal and the nasal border of the pocket in both cases. C denotes Cloquet canal; P denotes the PPVP.

Despite the presence of a clinical PVD with a Weiss ring, the vitreous cortex could be detected on the macula in 34 (40.5%) eyes (12 men, 22 women) of the 84 eyes (26 men, 58 women) with high myopia, and the incidence was significantly (P < 0.001) higher than in the controls (9 of 104 eyes, 8.7%). Among the 34 highly myopic eyes with residual vitreous cortex, the vitreous cortex was detached from the retina except for adhesion at the fovea or optic disc in 12 eyes (Fig. 5). The vitreous cortex adhered uniformly to the retina as an epiretinal membrane in the remaining 22 eyes (Fig. 6). 

Figure 5

 

Color fundus photograph and SS-OCT images in highly myopic eyes of a 74-year-old woman with −17.0 D of refractive error. According to the biomicroscopic findings and SS-OCT images, the stage of PVD was complete PVD with Weiss ring. Color fundus photograph showed myopic fundus with staphyloma and chorioretinal atrophy in the right eye (left). Swept-source OCT shows the residual vitreous cortex (arrows) in the horizontal scan thorough the fovea (right). *Denotes the blurred reflex in the vitreous cavity.

Figure 6

 

Swept-source OCT images showing foveoschisis and the posterior vitreous cortex. The case of the right eye of a 59-year-old woman with −14.0 D of refractive error (top). Swept-source OCT shows myopic foveoschisis, and the vitreous cortex (arrow) indicates a partial (perifoveal) PVD in the horizontal scan (top). The case of the right eye of a 56-year-old woman with −15.0 D of refractive error (bottom). In this case, we defined a complete PVD as detached posterior vitreous cortex with a Weiss ring by biomicroscopy. Swept-source OCT shows the vitreous cortex adhering to the retina as an epiretinal membrane (arrows) and myopic foveoschisis in the vertical scan.

Myopic foveoschisis developed in 14 (9.3%; 2 men, 12 women) of the 151 eyes (61 men, 90 women) with high myopia (Fig. 6); no foveoschisis was present in the 363 control eyes. Among the 14 eyes with foveoschisis, 3 (21.4%) had a partial PVD and 11 (78.6%) had a complete PVD; residual cortex was seen in 3 (27.3%) of 11 eyes with a complete PVD. There was a significant (P < 0.001) difference in the incidence of foveoschisis between men (2 of 61 eyes, 3.3%) and women (12 of 90 eyes, 13.3%). The mean refractive errors of the patients with foveoschisis were 9.0 ± 1.4 D in men and 13.0 ± 2.7 D in women. 

Discussion

This is the first description of precursor stages of PVD with SS-OCT in high myopia. Before development of a complete PVD with a Weiss ring, the PVD started in the paramacular area and progressed to a perifoveal PVD and vitreomacular separation. The precursor stage of PVD occurs at a younger age in high myopia. Swept-source OCT confirmed the early development of PVD in high myopia. The PPVPs were larger in high myopia as we previously reported. ¹⁷ The shapes of PPVPs alter with the position changes of the patients. ¹⁸ Though we performed the OCT for all cases in a seated position, the sizes of PPVPs may vary depending on not only the position of the patient but also eye movement, thus varying from day to day. 

The current study showed that the vitreous cortex frequently remains on the retina in highly myopic eyes despite a Weiss ring. Morita et al. ¹⁹ reported that vitreous liquefaction begins at a relatively young age in patients with high myopia and progresses with age and axial elongation from biomicroscopic findings. Yonemoto et al. ²⁰ reported that the higher the degree of myopia, the younger the age at which PVD begins. 

Posterior vitreous detachment occurs when liquefied fluid escapes through the disrupted vitreous cortex into the retrohyaloid space. The larger PPVPs reflect earlier vitreous liquefaction in high myopia, which may cause earlier partial and complete PVDs. The high incidence of residual vitreous cortex on the retina after complete PVD in high myopia is interesting. Another explanation for the residual cortex is splitting of the vitreous cortex. Because the vitreous cortex has a lamellar structure, ^13,21 the outermost layer of the cortex may remain on the retina during development of a complete PVD with Weiss ring. Because the posterior wall of PPVP at the fovea is very thin, residual cortex may exist that is not detected by SS-OCT. We defined the vitreous cortex remaining on the retina after occurrence of complete PVD with Weiss ring as the residual vitreous cortex. There is a possibility that some cases were classified as no PVD in spite of complete PVD if we overlooked Weiss ring in biomicroscopic findings. 

Foveoschisis is a common feature in severely myopic eyes with posterior staphyloma. ^5,22 It might be a progressive condition, and complications appear to be related to the presence of vitreoretinal tractions. ²² Baba et al. ²³ reported that the prevalence of foveoschisis without a macular hole was 9.0% in highly myopic eyes with posterior staphyloma. In the current study, the incidence of myopic foveoschisis was 9.3%. Among the 14 eyes with foveoschisis, three (21.4%) eyes had a partial PVD, which may cause anterior traction on the macula. However, 11 (78.6%) eyes had a complete PVD, and 8 (72.7%) of 11 eyes had no residual cortex. Thus, if a complete PVD is present in myopic foveoschisis, most cases had no vitreous traction from the residual cortex. The foveoschisis in highly myopic eyes resolved after vitrectomy that included a core vitrectomy, surgically induced PVD, removal of the premacular vitreous cortex, and an internal limiting membrane in the posterior staphyloma. ^24,25 The relative rigidity of the internal limiting membrane or sclerotic retinal vessels may provide anterior traction on the retina against the backward force exerted by the elongated axial length. ²⁶  

Although axial length information should be given to allow a valid judgment of the degree of myopia, we did not measured it in this study. Further study should be undertaken by measuring axial length using ultrasound biometry or partial coherence laser interferometry (IOL Master; Carl Zeiss, Jena, Germany). 

In conclusion, the current study showed the pathologic findings in the posterior vitreous in high myopia using SS-OCT. In addition to complete PVDs, partial PVDs around the fovea insidiously develop earlier in highly myopic eyes. The incidence of residual vitreous cortex with complete PVD was high in eyes with high myopia. In contrast to our expectations, there was no residual vitreous cortex in most patients with myopic foveoschisis. In a minority of patients, the residual cortex may play a role in the development of myopic foveoschisis. 

Acknowledgments

Disclosure: H. Itakura, None; S. Kishi, None; D. Li, None; K. Nitta, None; H. Akiyama, None