This observational study utilized OCT, OCTA, and microperimetry to analyze the structural, vascular, and functional effects of silicone oil endotamponade on the optic nerve and macula. To our knowledge, this is the first study to determine the correlation between all these factors, attempting to elucidate the frequently unexpected visual outcomes following successful retinal detachment surgeries.

All our patients had macula-off rhegmatogenous retinal detachment, with successful reattachment following PPV with SO. Subsequently, there was a significant improvement in vision following SO removal.

In our pilot study conducted in 2018, we employed microperimetry to evaluate retinal sensitivity both before and after the removal of SO. We aimed to investigate the impact of SO on the retina. Interestingly, our findings indicated that irrespective of the duration of SO tamponade and the removal of SO led to a significant improvement in overall retinal sensitivity. Additionally, the sensitivity at the middle, inner, and outer rings of the retina also showed improvement [16]. In the current study, there was also a significant increase in the retinal sensitivity of each ring (middle, inner, and outer) postoperatively (p-value < 0.001, < 0.001, and 0.002 respectively).

In our study, while the FAZ area did not change significantly, the CFT showed a significant increase post-SO removal. This finding is consistent with previous studies, which demonstrated that the decrease in central macular thickness due to SO removal following surgical treatment of RRD is reversible, and CFT increases after SO removal. The decrease in thickness can be attributed to various mechanisms, including the toxic impact on ILM, the mechanical impact on the layers of the inner retina, or the alteration of the retinal environment due to the oil’s hydrophobicity [3, 17, 18].

Although our study failed to identify a correlation between the FAZ area in RRD and BCVA, Woo et al.’s study showed that both the superficial and deep FAZ areas in cases with macula-off RRD were negatively associated with BCVA during the first 2 months after RRD repair. In addition, the deep FAZ area was larger in macula-off RD eyes compared to macula-on RRD eyes. They hypothesized that the DCP might be more susceptible to hypoxia and that the expansion of the deep FAZ area might be a sign of macular ischemia that is severe. However, Sato et al. and Hong et al. found no association between postoperative BCVA at 1, 3, and 6 months following vitrectomy in macula-off RRD and the superficial or deep FAZ area [10, 19].

The aforementioned study focused exclusively on the changes in the FAZ area during the period after vitrectomy and did not investigate the period following silicone oil (SO) removal. It is noteworthy that the FAZ area in our study did not demonstrate significant changes pre- and post-SO removal and did not exhibit any correlation with the BCVA. We can infer that the SO tamponade was not responsible for the changes that may have been observed in these studies.

OCTA allows non-invasive imaging of the macular and optic nerve perfusion. Given the unexplained changes in visual acuity changes post-SO removal that did not correlate with the significant improvement in retinal sensitivity, our study aimed to utilize OCTA to identify any potential contributing vascular changes.

In our case series, we found no significant changes in the macular VD. However, we observed a significant increase in the postoperative vascular density of the whole image and peripapillary radial capillary plexus (p-value < 0.001 and 0.002), respectively, following SO removal.

Numerous studies have revealed that SO tamponade leads to both a reduction in outer retinal thickness and VD [20,21,22]. Nevertheless, few studies have evaluated the effect of SO removal on macular perfusion [23, 24]. In Prasuhn et al. [23], SO removal did not affect retinal perfusion values, with only a significant increase in the CCP postoperatively. Lee et al. reported no significant difference in VD following SO removal and 6 months following. Bayraktar et al. conducted an OCTA study to examine the superficial and DCP of the retina before and after SO removal. They found that the values of VD remained consistent throughout the study period [25].

With regard to the significant increase in the VD of the whole optic nerve image and in peripapillary radial plexus following SO removal, Wang et al. also detected an elevation in peripapillary capillary density following SO removal (most noted in the superior hemifield). They suggested that the use of SO tamponade could potentially impact peripapillary blood flow through capillary compression [26]. In our study, despite the strict exclusion of patients with increased IOP, a significant decrease in IOP following SO removal was still observed, which may have contributed to improved optic nerve perfusion. This finding aligns with the study conducted by Chen and colleagues, who concluded that peripapillary VD elevated following the decrease in IOP. However, they reported only mild changes in IOP, which did not influence macular microvascular parameters. It is worth noting that their study is different from ours as it was conducted on patients with ocular hypertension [27].

In our study, a substantial improvement in BCVA was observed following SO removal. However, no significant correlations were identified between BCVA and any of the microperimetry or OCTA data, except for VD of the fovea in SCP preoperatively. Despite a significant increase in retinal sensitivity, no significant correlations were found between preoperative, postoperative, or percentage change in retinal sensitivity and macular vascular density.

In order to eliminate confounding factors, none of our patients had associated cataract extraction during the SO removal. Moreover, any patient who developed high IOP with optic neuropathy was excluded.

The improved retinal sensitivity on microperimetry in our pilot research, as well as in another investigation comparing SO with gas tamponade in cases of retinal detachment, did not show a significant correlation with BCVA, which also exhibited a notable improvement. The potential impact of SO tamponade on ocular functioning, particularly in resolving spatial patterns for sensitivity and acuity in microperimetry, was utilized to elucidate this finding [16, 18].

Other studies also correlated macular VD in eyes undergoing PPV with SO tamponade and BCVA and did not report significant correlations. [22, 28, 29]. Therefore, we can conclude that changes in visual acuity following anatomically successful PPV are not likely attributed to a potential deleterious effect of SO tamponade. However, attributing the improvement in BCVA and retinal sensitivity solely to the SO removal is not completely valid, as it may be one of several factors, such as the natural healing process following anatomical reattachment of the retina, that contributed to this finding.

Wang et al. [11] and Hong et al. [19] also could not find significant correlations with the SCP or DCP VD. Nevertheless, they found a significant correlation with CCP VD, suggesting that the restoration of the choriocapillaris and deep retinal plexus is substantially associated with functional consequences. It is important to note that, unlike our study, these studies only evaluated patients during SO tamponade and did not follow-up with patients.

The duration of SO tamponade did not demonstrate any significant correlation with retinal sensitivity or vascular density of OCTA macular layers and optic nerve head parameters, which is consistent with findings from previous studies [23, 30, 31]. Lee et al. [24] found a significant association between the duration of SO tamponade and DCP VD. They hypothesized that the detrimental effects of SO on retinal tissues caused a drop in macular VDs at a particular time point following surgery [32].

In our study, the duration of silicone oil (SO) tamponade varied from 3 to 11 months, with a mean of 6 months, and yet, no correlation was found with best-corrected visual acuity (BCVA), allowing us to dismiss the potentially harmful effect related to this factor.

Furthermore, the total retinal sensitivity was positively and significantly linked with CCP (r = 0.395, p-value = 0.031). The choroid is considered the principal source of oxygen to the retina, and thus, altering its circulation can affect both the RPE and outer retina, leading to a disturbance in visual function [33]. According to Prasuhn et al., this substructure should not be disregarded when examining morphological changes [23].

The reason behind the increase in retinal sensitivity upon removal of the silicone oil remains unclear. Adverse effects such as cataracts, silicone emulsification, and secondary glaucoma can worsen after the removal of SO. These effects might significantly influence vision and retinal sensitivity. Nevertheless, we categorically excluded these individuals from our investigation.