Intravitreal Injections for Macular Edema in Silicone Oil Filled Eyes

Introduction

Silicone oil (SiO) is a commonly used intraocular tamponade in complicated vitreoretinal surgeries. It maintains the adhesions between the neurosensory retina and the retinal pigment epithelium (RPE), especially after complicated retinal detachments due to proliferative vitreoretinopathy (PVR), tractional detachment caused by proliferative diabetic retinopathy (PDR), viral retinitis, giant retinal tears, or trauma.^1,2

SiO has been reported to cause retinal changes such as cystoid macular edema (CME), submacular fluid, macular epiretinal membrane (ERM) and undulated inner retinal layer.³ The incidence of post-vitrectomy macular edema may range from 6% to 36%.^4–7 Macular edema (ME), either vitrectomy-related or SiO-induced, may involve increased growth and inflammatory factors, which have been reported in retro-SiO fluid studies, including interleukin 6 (IL-6) and LDH.⁸ Other causes may be due to underlying diseases such as retinal vein occlusion (RVO) and diabetic retinopathy (DR).^9–13

In complicated cases, when SiO removal may lead to recurrent retinal detachment, it is preferable to keep the SiO tamponade for an extended period to ensure retinal attachment and stabilization.^14,15 Unfortunately, these eyes may be more frequently prone to develop macular edema.¹⁶ There are several potential treatment options available to address post-vitrectomy macular edema.^17–20 These treatments include topical and systemic nonsteroidal inflammatory drugs (NSAIDs), intravitreal anti-vascular endothelial growth factor (anti-VEGF) injections, and periocular, intravitreal, or systemic steroids. In cases of refractory macular edema after topical treatment or caused by co-existent retinal pathologies, intravitreal injections are the mainstay of treatment.²¹

The presence of SiO tamponade may affect the delivery and concentration of drugs injected into the vitreous cavity, and hence their effectiveness is questionable.^22,23 Limited research has been conducted on treating macular edema in silicone oil-filled eyes by intravitreal SiO injections. A few studies have shown that intravitreal SiO injections of anti-VEGF, steroids, or other therapeutic agents are effective for treating macular edema and preserving vision (Table 1).^22–37 Intra-SiO injections were not associated with special complications and were deemed to be safe. Nevertheless, there is still a lack of consensus regarding their efficacy, which restricts their application on clinical grounds. Consequently, we aim to share our experience in using intra-SiO injections for treating postoperative macular edema in patients after complicated retinal detachment repair surgery who were left with silicone oil tamponade.

Table 1 Review Table

Methods

This is a retrospective case series study from a single referral center. The study was conducted in accordance with the tenets of the Declaration of Helsinki and was approved by the institutional ethics committee of the Hadassah-Hebrew University Medical Center.

Eight eyes (eight patients), that had undergone a complicated retinal detachment surgical repair with silicone oil tamponade and had developed macular edema during follow-up visits, were included in the study. Macular edema, including intraretinal and subretinal fluid, was diagnosed based on Optical Coherence Tomography (OCT) imaging using Spectral Domain-OCT (Heidelberg Engineering, Heidelberg, Germany).^38,39

The data collected included age, gender, previous medical and ocular history, type of intravitreal injections, number of intravitreal injections, treatment regimen, follow-up period, visual acuity (VA), central subfield macular thickness (CSMT) and features of the retinal fluid (intraretinal or subretinal fluid) based on OCT images. VA and CSMT values were obtained before the first intravitreal injection, before every subsequent injection and at the last follow-up.

All values of VA were converted to LogMAR. To provide numerical values for low visual acuities, the following conversions were made: LP (light perception) = 4 (Snellen equivalent 0.0001), HM (hand movement) = 3 (Snellen equivalent: 0.001), FC (finger counting) = 2 (Snellen equivalent: 0.01).⁴⁰

Intravitreal injections were performed according to standard protocols using a 1 mm Luer lock syringe and 30-gauge needle that was inserted for 3–4 mm at the inferior quadrants. The injections were performed slowly to avoid mixing the drug with the silicone oil bubble. Intravitreal injections included Triamcinolone Acetonide (TA, 4mg/0.1cc), sustained-release dexamethasone intravitreal implant (Ozurdex, 0.7 mg) and anti-VEGF (Bevacizumab 1.25mg/0.05cc, Aflibercept 2mg/0.05cc).

Statistical Analysis

Data was recorded in Microsoft Excel and analyzed using GraphPad Prism 7.0. The t-test was used for statistical significance testing of interval scale parameters across two groups. For statistical analysis of VA in LogMAR or CSMT, the Wilcoxon signed rank test was used. The threshold for statistical significance was defined as P-value < 0.05.

Results

Eight eyes of eight patients were included in the study (5 males and 3 females). The mean age (±SD) of the patients was 64.75±8.9 years. All patients had undergone previous complicated retinal detachment surgery and were left with silicone oil tamponade. All eyes were pseudophakic except patient number 6 (Table 2). They all had similar type of SiO, 1300cs (centistokes), except patient #2 who had 5000cs SiO tamponade, and patient #7 who had Oxane^® 5700cs heavy SiO (Table 2).

Table 2 Basic Characteristics of the Patients Included in the Study

All eyes developed macular edema during their follow-up visits. The mean time (±SD) elapsed between retinal detachment surgery and the development of macular edema was 9±16.5 months (range 0.2–49.3 months). Intraretinal fluid (IRF) was the main component of this edema in all patients. In addition, patient #1 had submacular hemorrhage and patient #2 had subretinal fluid (SRF) (Table 2).

The mean follow-up period (±SD) was 37.2±26.4 months (range 9.6–85.2 months). The mean number of injections throughout the study period was 8.25±7.24 (median 6, range 1–20). The mean number of injections per year was 3.15±2.94. The number and type of injections that each patient received are summarized in Table 2. Eyes that received anti-VEGF injections were treated by the Treat and Extend (T&E) protocol while patients who received steroid injections were treated by the Pro-Re-Nata (PRN) protocol. During the study period, patient #3 underwent SiO removal surgery 9 months after the RD repair surgery. Patient #6 underwent SiO removal surgery 6 months after the RD repair surgery, and this was combined with cataract extraction due to visually significant posterior subcapsular cataract. Five patients received topical anti-inflammatory treatment before proceeding to intravitreal injections when macular edema was suspected to have an inflammatory component (Irvine-Gass Syndrome, Table 2).

The mean (±SD) visual acuity (VA, in LogMAR) and central subfield macular thickness (CSMT) before injections were 1.7±0.8 and 488±104 μm, respectively (Table 3). At the last follow-up visit, the mean (±SD) VA (LogMAR) and CSMT were 1.4±0.7 and 396±184 μm, respectively. In a comparison between VA and CSMT before injections and at the last follow-up, there was no statistically significant difference (p-value=0.45 and 0.11, respectively), but there was a clear trend of improvement (Table 3). During the treatment period, the eyes showed, in general, a fair response without significant worsening of the macular edema and VA (Figures 1, 2 and 3). Even though patients #3 and #6 underwent silicone oil removal surgery, the macular edema persisted, and they continued treatment with anti-VEGF injections afterwards.

Table 3 Clinical Outcomes

Figure 1 Central Subfield Macular Thickness (CSMT, in microns) graph for all patients (P. 1 to 8) at the day of receiving injection along the study period. On each point, the type of injection is represented by a sign as shown on the right side of the figure.

Figure 2 Visual acuity graph in LogMAR for all patients (P. 1 to 8) at the day of receiving injection along the study period. On each point, the type of injection is represented by a sign as shown on the right side of the figure.

Figure 3 B-mode OCT horizontal section scans of two patients. (A) Patient #3 before receiving intra-silicone oil injection of Ozurdex. (B) Patient #4 before receiving intra-silicone oil anti-VEGF (Aflibercept) injection. (C) Patient #3 four months after receiving intra-silicone oil Ozurdex injection, showing improvement of the macular edema. (D) Patient #4 three months after receiving two intra-silicone oil anti-VEGF (Aflibercept) injections with 6 weeks interval between both injections, showing improvement of the macular edema.

Two out of eight patients experienced side effects. Patient #3 experienced partial emulsification of the SiO 3 months following intravitreal Ozurdex injection and the SiO was extracted 9 months after the initial RD repair surgery. Patient #6 had progression of posterior subcapsular cataract 3 months after intravitreal TA injection. The SiO was removed in combination with cataract extraction, 6 months post the initial RD repair surgery, with improved vision during follow-up visits, postoperatively. None of the patients experienced elevated intraocular pressure, severe vision loss, retinal detachment, or endophthalmitis.

Discussion

Silicone oil (SiO) filled eyes may develop macular edema postoperatively.⁴¹ Although macular edema may be related to underlying diseases, it may also be silicone oil-induced or vitrectomy-related, as underlying inflammation has been shown to play a role. The incidence of macular edema after vitrectomy with SiO tamponade may range between 6% to 36% as reported by multiple studies.⁴ In addition, a longer duration of silicone oil tamponade may increase the risk of developing macular edema.^3,16,41–43 In our study, macular edema developed, on average, 9 months postoperatively (range 0.2–49.3 months). Yang et al reported an earlier incidence, 5.7 months after surgery.¹⁶

The vitreous cavity has been increasingly used as a reservoir for drugs to treat intraocular diseases.⁴⁴ Any intraocular tamponade may alter the elimination rate of drugs from the vitreous cavity.⁴⁵ Therefore, the presence of SiO can affect drug pharmacokinetics and pharmacodynamics, which may interfere with the macular edema response to intravitreal injections. Previous studies reported on using intravitreal injections of anti-VEGF or steroids for treating macular edema in patients after vitrectomy and SiO tamponade.^27,28 Several studies showed promising results and good response to treatment,^23,29 however, others concluded that SiO may not be a good candidate as a drug reservoir (Table 1).^22,46,47 Unlike other studies which reported on using mainly one type of intra-SiO injection during their investigation, here we present a case series using repetitive multiple types of injections, in the same eye if needed, as well, including different Anti-VEGF agents and steroids (Table 2).

In our study, VA and CSMT did not significantly improve after intra-SiO injections, as was reported by others.^32,34,36 Nevertheless, there was a trend of improvement in both aspects (Table 3, Figures 1, 2 and 3). The insignificance may be attributed to the small number of participants and further studies with bigger samples are needed. However, leaving patients with untreated macular edema may cause irreversible deterioration of vision and intractable macular damage.^48,49 Therefore, treating these complicated cases, at least in order to preserve vision and prevent the worsening of macular edema, is valuable and worth considering.

In terms of side effects of intra-SiO injections, none of our patients experienced endophthalmitis or severe vision loss. There were no cases of elevated intraocular pressure, visual disturbances, or destabilization of the retinal attachment status. Special complications were absent as reported by other studies (Table 1). Emulsification of SiO was observed only in one patient (patient 3) three months after Ozurdex injection, and the SiO was removed shortly afterwards. However, it is unclear whether the Ozurdex injection promoted SiO emulsification, or if it is the natural course of the SiO lifetime. All patients were pseudophakic, except one patient (patient #6) who developed a visually significant posterior subcapsular cataract three months after intravitreal TA injection, and hence SiO was removed in combination with cataract extraction. Even though both patients underwent silicone oil removal surgery, the macular edema persisted, and they continued treatment with intravitreal anti-VEGF injections thereafter. Hence, the macular edema was mainly due to their underlying retinal disease and less likely vitrectomy-related or SiO-induced.

Regarding patients who received intra-SiO TA injections, TA was observed accumulating in the inferior vitreous cavity under the SiO during their follow-up visits (Figure 4). TA injection was performed using a 30-gauge needle, according to a standard protocol similar to anti-VEGF injections. Then, TA was injected slowly, in order to lower the risk of TA mixture with the SiO. Spitzer et al demonstrated that when TA is injected into SiO-filled eyes, biologically active concentrations are present only below the oil bubble and no TA is retained in the bubble, possibly because silicone oil has a lower specific weight than TA preparations. Furthermore, high local concentrations of TA crystals may be present on the retinal surface because of sedimentation. This sediment may interfere with vision when it is present in the macular area.³⁵ Nevertheless, none of our patients complained about floaters or decrease of vision related to the presence of TA in the vitreous cavity.

Figure 4 A Fundus picture focusing on the inferior retina of the left eye of patient 4, showing whitish material inferiorly underneath the silicone oil tamponade representing TA accumulation (the red circle).

Accumulation of drug substance in the inferior meniscus of the fluid underneath the SiO may be gravity-driven and was reported by Falavarjani et al while treating neovascular glaucoma (NVG), after vitrectomy and SiO endotamponade, with intravitreal anti-VEGF (Bevacizumab) injections.²³ In our study, there were no reports of anti-VEGF substance accumulation under the SiO. Interestingly, Falavarjani et al injected 2.5mg bevacizumab instead of the more commonly used dose of 1.25 mg, with good clinical response and no signs of retinal toxicity. Consequently, in patients treated with intra-SiO 1.25mg Bevacizumab injections, higher doses may be considered for persistent macular edema. However, future studies are required to further validate this issue.

Certain limitations should be considered in our study, including its retrospective nature, a small number of participants, a relatively short period of follow-up for some of the patients and a small number of injections. Best-corrected visual acuity was not available, and the measured Snellen visual acuity was converted to LogMAR. In addition, multiple patients were treated with additional anti-inflammatory topical drops and their effect on macular edema cannot be analyzed separately. Nonetheless, they were used mostly as a trial to postpone or obviate the need for intravitreal injections.

Conclusions

SiO-filled eyes may develop macular edema postoperatively that requires treatment. Based on our case series study, intra-SiO injections of anti-VEGF or steroids were safe and relatively effective. Both types were not associated with serious complications. Consequently, macular edema in silicone oil-filled eyes may be safely and effectively treated with intravitreal injections, to halt deterioration and preserve or improve vision, especially when SiO tamponade removal is not on the horizon.