Quantitative assessments of PVD using OCT are scarce. This study found that in eyes without pathologies, partial PVD can be observed even at age ≤ 10 years, and the PVD stage progressed with age. Furthermore, there was no significant difference in PVD stage between the left and right eyes in the same patient.
PVD was historically considered as an acute event occurring due to an abrupt break in the posterior vitreous cortex leading to mechanical collapse and separation of the degenerate vitreous away from the retina.1 However, OCT has shown that PVD progression is of a more chronic nature rather than an acute event [10, 11]. Previous studies using 7-mm and 9-mm time-domain OCTs have reported that partial PVD starts in the thirties in healthy subjects [6, 12]. However, the results of our study showed partial PVD even in children as young as 5 years old, indicating that the initial onset of PVD is much earlier than that reported.
This discrepancy could be due to the following reasons: first is the difference in the age of the subjects enrolled. Previous reports have examined healthy adult eyes, and few have examined PVD in normal eyes of minors using OCT [6, 8, 12]. It is assumed that minors were not enrolled due to the long testing time of OCT in these studies. Given that the SS-OCT device used in this study had a shorter acquisition time than the conventional SD-OCT device or montaged OCT method requiring multiple OCT images, it was possible to enroll minors, including children who have difficulty with prolonged positioning. The second factor is the imaging range. In a study using 25–36-mm OCT images montaged from three OCT images, 34 of 36 healthy subjects in their 20s had partial PVD [8]. This result suggests that PVD begins at a younger age than previously reported.
Moreover, in a study comparing 6-mm and 16.5-mm OCT images in the same eye, seven eyes were categorized as no PVD on 6-mm OCT, but these were upgraded to partial PVD on 16.5-mm OCT [13]. The study concluded that 6-mm OCT scans cannot detect early partial PVD. Considering these previous findings, we hypothesized that the earlier age onset of stage 1 PVD in our study was because we were able to identify peripheral partial PVD by using widefield OCT images. To confirm this hypothesis, we additionally examined the PVD stage on a 6-mm OCT image centered on the fovea taken from the 23-mm OCT image used in this study. On the 6-mm OCT images, 85.7%, 56.5%, 44.7%, 25.0%, 10.0%, 17.4%, and 5.0% of those aged < 10 years, 10–19 years, 20–29 years, 30–39 years, 40–49 years, 50–59 years, and 60–69 years displayed full vitreoretinal adhesion in the macula and appeared to show no PVD (stage 0). However, these patients were found to have stage 1 partial PVD on the original 23-mm widefield OCT image (Supplemental Fig. 1) as in the representative case in Supplemental Fig. 2.
To minimize the effect of individual patient characteristics on the statistical analysis, SS-OCT images were obtained for bilateral eyes in all subjects, but only the eyes of those with advanced PVD stage were first evaluated. Then, we examined the bilateral eyes and found that the rate of PVD progression was identical in both left and right eyes. There was no significant difference in the PVD stage progression between the left and right eyes. To our best knowledge, this is the first study to evaluate and report that there is no difference in PVD progression between the left and right eyes in the same patients.
Sex differences in the incidence of ocular diseases have been reported, such as a higher incidence of rhegmatogenous retinal detachment in males than in females [14] and a higher incidence of macular hole in females than in males [15]. However, there have been no reports on the individual incidence of PVD in the left and right eyes. The lack of significant differences in the longitudinal PVD progression between the left and right eyes may explain the similar incidence of PVD-related rhegmatogenous retinal detachment and macular hole between the left and right eyes.
Although the PVD stage was concordant in the two eyes in 80% of the subjects, there were few subjects who had concordant stage 2 or stage 3 PVD. In addition, 8 subjects (3.7%) had three different PVD stages, such as stage 1 on the right eye and stage 4 on the left eye. Although the reason for this difference is unclear, it should be noted that assessment of only one eye may lead to under- or overestimation of the PVD stage. In clinical practice, patients with macular hole in one eye should be evaluated for PVD in the other eye, and if that eye does not show fovea detachment, i.e., PVD stage 2 or lower, and are recommended for regular follow-up.
Several limitations of this study should be acknowledged. First, it was difficult to assess PVD in patients with vitreoschisis. In a study using scanning electron microscopic observations, 44% of eyes with spontaneous PVD showed vitreous cortex remnants on the retinal surface [16]. A conventional SD-OCT study also detected vitreoschisis in 53.3% of eyes with macular holes and 43.2% of eyes with macular pucker [17]. In a study using montaged OCT, vitreoschisis was prevalent in 41.2% of eyes without vitreous-retina-choroid pathologies.8 It is difficult to distinguish the vitreoschisis line from the posterior vitreous line. This may explain the higher incidence of stage 1 PVD in this study than in previous reports. SD-OCT uses a longer OCT wavelength and thus has a higher resolution than SS-OCT [18]. Future research should examine the partial PVD sites in more detail by using SD-OCT. Second, although age, sex, and refraction have been reported to be associated with PVD [19–21], the sex ratios and refractions differed among the age groups in this study. However, in an additional analysis, there was no significant difference in the mean age of males and females (41.9 ± 24.4 vs. 42.2 ± 24.2, P = 0.8824), nor was there a significant difference in the PVD stage between the sexes (2.3 ± 1.5 vs. 2.2 ± 1.5, P = 0.7262). In addition, only subjects with a refractive power of + 3.0 diopters to -5.0 diopters were included. Thus, we believe that the effects of sex and refractive power were minimized. Finally, 10.1% of the subjects had pseudophakic eyes. Studies using SD-OCT have shown that cataract surgery accelerates PVD [22]. To more accurately assess the natural history of PVD, we need to evaluate only phakic eyes even in the elderly population.
Despite these limitations, we minimized the influence of individual characteristics and enrolled a large sample size. Importantly, we also evaluated PVD quantitatively. In previous OCT studies, PVD was qualitatively assessed. Further, paramacular and perifoveal PVD were defined as stage 1 and 2, respectively, thus limiting the reproducibility of the PVD staging system. Meanwhile, the staging system used in this study can be conveniently applied staging across different OCT devices and graders, thus making it a suitable method in future studies of OCT for PVD.
In conclusion, partial PVD in eyes without pathologies occurs at an earlier age than previously reported, as evidenced by the detection of PVD even in subjects aged < 10 years. The PVD stage progresses with age, but the rate is concordant between the left and right eyes. These findings can be useful for future widefield OCT studies on vitreoretinal interface diseases.