The impact of removing the epiretinal membrane and inner limiting membrane for sustained subretinal fluid by macular neovascularization refractory to anti-VEGF therapy

Abstract

Purpose

Anti-vascular endothelial growth factor (VEGF) therapy is the most prevalent intervention for exudative lesions secondary to neovascular age-related macular degeneration (nAMD) and other macular neovascularization (MNV). However, in some cases refractory to the latest anti-VEGF agents is associated with epiretinal membrane (ERM) or vitreomacular traction. We applied a vitrectomy to remove those pathologies which may be effective for reducing the exudation.

Observations

In this case report, we present 2 cases with sustained subretinal fluid and macular neovascularization secondary to nAMD or dome-shaped macula that poorly responded to anti-VEGF therapy. In both cases, removing thin ERM or vitreomacular traction with an inner limiting membrane peeling promptly resolved the subretinal fluid and no recurrence was observed thereafter.

Conclusions and importance

Vitrectomy could be an effective modality for anti-VEGF drug-resistant MNV cases with vitreomacular traction or ERM even in the anti-VEGF era.

1. Introduction

Anti-vascular endothelial growth factor (VEGF) therapy is the most used treatment for neovascular diseases at the fundus.1, 2, 3 Anti-VEGF therapy promptly resolves exudative change caused by macular neovascularization (MNV) and vascular hyperpermeability. However, some cases do not respond to anti-VEGF therapy.⁴ Several reasons include epiretinal membrane (ERM) or vitreomacular traction.5, 6, 7 Schulze et al. mentioned the merit of vitrectomy as an alternative intervention for cases refractory to anti-VEGF therapy.⁶ Subsequently, some new anti-VEGF agents have been developed which have a stronger ability to inhibit VEGF or co-inhibit dual angiogenic molecules to suppress MNV. Hence, the current discussion about MNV control is focused on switching drugs rather than surgical interventions. However, there are still some cases that poorly respond to the latest anti-VEGF agents and the removal of vitreomacular traction or ERM is effective for them. In this case report, we present 2 cases with subretinal fluid (SRF) secondary to MNV that were refractory to repeated anti-VEGF therapy. Removing thin ERM and inner limiting membrane (ILM) effectively resolved the SRF in those cases.

2. Cases report

Case 1

A 74-year-old male was referred to our hospital for neovascular age-related macular degeneration (nAMD) in the right eye refractory to repeated anti-VEGF therapies. He had received 3 injections of ranibizumab and 15 injections of aflibercept in the right eye over 10 years with a pro re nata (PRN) regimen and the MNV was well controlled during this period. However, it became refractory to aflibercept accompanied by ERM development. In the latest recurrence, aflibercept was injected every 3 months. He had chronic renal failure and received dialysis. He was also taking medications for diabetes mellitus and systemic hypertension. At the first presentation to our hospital, his right eye's best corrected visual acuity (BCVA) was 0.15 in decimal fractions. Although his fundus was not clearly visible due to a cataract, optical coherence tomography (OCT) showed submacular neovascularization accompanied by SRF and thin ERM (Fig. 1). Fluorescein angiography (FA) and indocyanine-green angiography (ICGA) revealed type 2 macular neovascularization (MNV) in his right eye.

Three monthly intravitreal injections of brolucizumab were performed, but there was no change or even increase in the SRF after the treatments. Therefore, he underwent a pars plana vitrectomy combined with cataract surgery. After phacoemulsification and intraocular lens implantation, a 3-port 27-gauge vitrectomy was performed using CONSTELLATION ® Vision System (Alcon Japan Ltd, Tokyo, Japan). Subtotal vitrectomy was done and the ERM and ILM at the macula were removed using visualization with triamcinolone acetonide and ICG (Fig. 2). After confirming the complete removal of ILM and ERM by intra-operative OCT, fluid-air exchange was added. Ten days after the surgery, the subretinal fluid was completely resolved, but the subfoveal neovascular lesion remained (Fig. 2). Anti-VEGF therapy was re-started with faricimab with a treat & extend regimen begun from 8-week intervals because faricimab became available from that point. After 7 injections, the subfoveal neovascular lesion was mostly resolved and the injection interval was extended to 10 weeks. No recurrence occurred until a year and a half after the surgery and his right eye's BCVA was improved to 0.4.

Case 2

A 57-year-old female was referred to our hospital for nAMD in the right eye. Her right eye's BCVA was 0.3. FA/IA and OCT detected a thin type-1 MNV accompanied by SRF at the macula. A high myopia (refractive error ≒ −10D, axial length ≒ 27mm) and dome-shaped macula were seen in both eyes. The case was diagnosed as MNV associated with a dome-shaped macula. Two monthly intravitreal injections of aflibercept were applied and the SRF was resolved, but her BCVA did not change. Two years later, SRF recurred with sustained type-1 MNV (Fig. 3). Her right eye's BCVA was 0.5. OCT revealed an adhesion of the posterior vitreous membrane with mild traction at the fovea (Fig. 4). An additional injection of aflibercept was performed and SRF decreased but remained after a month. Then, a 3-port 27-gauge vitrectomy with cataract surgery was conducted. Subtotal vitrectomy was done and the ERM and ILM at the macula were removed using visualization with triamcinolone acetonide and ICG. No fluid-air exchange was performed. Ten days after the surgery, the subretinal fluid was completely resolved (Fig. 4), and no recurrence of SRF was observed for over 5 years. Her right eye's BCVA was improved to 0.8.

Fig. 1.

Clinical findings of Case 1 at the first presentation to our hospital. (A) optical coherence tomography (OCT) showed subretinal hyperreflective foci and subretinal fluid (SRF) (asterisks) accompanied by a thin epiretinal membrane (ERM) (arrows). (B) Fluorescein angiography detected a classic choroidal neovascularization. (C) Indocyanine-green angiography revealed macular neovascularization. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 2.

Fundus images during vitrectomy and chronological changes in OCT findings. (A) ERM was visualized with triamcinolone acetonide and removed. (B) The inner limiting membrane at the macula was peeled off after staining with indocyanine-green. (C) After 3 monthly injections of brolucizumab. (D) Ten days after the vitrectomy. (E) A year and a half after the vitrectomy.

Fig. 3.

Clinical findings of Case 2 when subretinal fluid was recurred. (A) Fluorescein angiography detected an occult choroidal neovascularization. (B) Indocyanine-green angiography revealed faint macular neovascularization (arrowheads). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 4.

Chronological changes in OCT findings. The left: horizontal images, The right: vertical images. (A) At the recurrence of SRF. Thin ERM (arrow), small retinal pigment epithelial detachment (small arrows), and vitreomacular traction (arrowheads) are detected. (B) A month after the injection of aflibercept. (C) Ten days after the vitrectomy.

3. Discussion

The first-line treatment for MNV secondary to nAMD or other causes is anti-VEGF therapy in the current clinical practices.³ However, 20–30 % of treatment-naïve cases are refractory to anti-VEGF therapy.⁴ The presence of ERM or vitreomacular traction could be a reason for the poor response and removal of them may retrieve the reactivity to anti-VEGF therapy.⁸ It is still in discussion why the removal of ERM and vitreomacular traction benefits the treatment for an exudation associated with MNV. Vitreomacular adhesion may cause tangential forces on the macula and vitreomacular shear forces, exerting mechanical traction onto the retina, or by continuous mechanical traction, which in turn stimulates the proliferation of retinal cells and the build-up of ERM which exerts tangential forces on fovea.⁶ However, some vitreomacular traction and ERMs are not distinct and unlikely to affect the structure of the macula. Kimura et al. suggested three possible mechanisms for the effect of the ERM and ILM removal on resolving SRF.⁸ First, releasing the tractional forces on the ILM may suppress the secretion of signaling factors from Müller cells, thus downregulating chronic inflammation in the eye. Second, an adherent vitreous membrane over the macula may prevent the effective clearance of VEGF and other cytokines into the vitreous cavity.⁹ Releasing adherent vitreous from the macula may enable the diffusion of VEGF and other cytokines into the vitreous cavity and increase the clearance of molecules, including VEGF.¹⁰ Third, vitrectomy might increase the intraocular oxygen tension, leading to the suppression of both hypoxia-inducible factor 1 and VEGF production.¹¹ In the present case report, Cases 1 and 2 showed a thin ERM and mild vitreomacular traction, respectively. In particular, Case 1 did not respond to brolucizumab, the strongest anti-VEGF agent in the current clinical use, but anti-VEGF therapy with faricimab was effective after the ERM removal, which suggested a tangential force at the posterior pole retina that resisted anti-VEGF therapy. In Case 2, the absence of recurrence after removing vitreomacular traction and ILM suggested better clearance of VEGF after the surgery. In our clinical practices, we usually start treatment with anti-VEGF therapy for MNV even if the case is accompanied by ERM or vitreomacular traction since it is difficult to precisely assess the influence of those pathologies on the exudative change in each MNV case before treatments and anti-VEGF therapy is believed to be safer than vitrectomy as an initial treatment. However, if the initial treatment with anti-VEGF fails possibly due to the ERM or vitreomacular traction, surgical treatments should be considered as an alternative treatment option. On the other hand, vitreous removal may accelerate intraocular clearance and reduce the concentrations of anti-VEGF agents if required subsequently which might result in the need for more frequent injections. Currently, the indication of vitrectomy for MNV is case-sensitive since there are no established criteria for surgical intervention. Further discussions are needed to determine certain indications of vitrectomy for MNV refractory to anti-VEGF therapy.

In conclusion, vitrectomy could be an effective modality for anti-VEGF drug-resistant MNV cases with vitreomacular traction or ERM and hence should be positively considered even in the anti-VEGF era.

Funding sources

This manuscript did not receive any funding.

Data availability statement

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.

CRediT authorship contribution statement

Aki Nihei: Writing – original draft, Investigation, Data curation, Conceptualization. Manabu Yamamoto: Writing – review & editing, Validation, Formal analysis, Data curation. Kumiko Hirayama: Writing – review & editing, Validation, Data curation. Akika Kyo: Writing – review & editing, Validation, Data curation. Norihiko Misawa: Writing – review & editing, Validation, Data curation. Takeya Kohno: Writing – review & editing, Validation, Data curation. Shigeru Honda: Writing – review & editing, Supervision, Investigation, Formal analysis, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.