Efficacy of a simple intravitreal perfluoropropane injection in treating unclosed idiopathic macular holes following vitrectomy

Abstract

Background

This study aimed to evaluate the efficacy of a simple intravitreal injection of perfluoropropane (C₃F₈) in treating unclosed idiopathic macular holes (IMHs) in patients who had previously undergone primary pars plana vitrectomy (PPV).

Methods

This study was a retrospective, observational clinical study. It included patients diagnosed with unclosed IMHs following primary PPV combined with internal limiting membrane peeling and air tamponade. Optical coherence tomography (OCT) performed 1 week after PPV revealed unclosed IMHs with the presence of the ‘cuff’ sign and intraretinal cysts. The following day, these patients received a simple intravitreal C₃F₈ injection. Comprehensive evaluations, including best-corrected visual acuity (BCVA), indirect ophthalmoscopy, fundus photography, and OCT, were performed before PPV, 1 week after surgery, and at follow-up intervals of 1–3 months after the intravitreal gas injection.

Results

The minimum linear diameter (MLD) of the macular holes (MHs) 1 week before C₃F₈ injection was 335.1 ± 74.3 μm. Following C₃F₈ tamponade, the closure rate of the MHs was 100%. The mean BCVA before C₃F₈ tamponade was 0.68 ± 0.17 logMAR (20/100) after primary PPV, which improved to 0.48 ± 0.19 logMAR (20/63) after C₃F₈ tamponade, showing a statistically significant difference (P = 0.01).

Conclusions

For patients with unclosed MHs after primary PPV, the presence of the ‘cuff’ sign on OCT suggests that retreatment can be effectively achieved through a simple intravitreal gas injection. This approach is straightforward, practical, and effective.

Background

Pars plana vitrectomy (PPV), internal limiting membrane peeling (ILMP), and intravitreal air or gas injection are the primary surgical procedures for treating idiopathic macular holes (IMHs) [1]. While the primary success rate of these procedures is approximately 90%, some patients still experience MHs that fail to close after surgery [2]. In cases where MH closure is not achieved after primary PPV, reoperation can result in a closure rate as high as 85%, and partial improvement in visual acuity [3]. The standard approach for reoperation involves an extended ILMP and other auxiliary methods, which are often complex and expensive. In contrast, we treated a group of patients with unclosed MHs after PPV using a simple intravitreal gas injection. This approach is straightforward and effective.

Methods

This retrospective, observational clinical study was approved by the Ethics Committee of Tianjin Medical University Eye Hospital (Approval number: 2023KY(L)-25) and adhered to the Declaration of Helsinki. Written informed consent was obtained from all participants.

The study included eight patients (eight eyes) who underwent PPV combined with ILMP and air tamponade for treating IMHs at Tianjin Medical University Eye Hospital between January 2016 and September 2023. The inclusion criteria were as follows: (1) patients who underwent PPV combined with ILMP and intravitreal gas injection for IMHs, and the MH remained unclosed 1 week postoperatively; and (2) optical coherence tomography (OCT) after primary PPV revealed an upturned hole edge, subretinal fluid beneath the hole edge (the ‘cuff’ sign), and intraretinal cysts. The exclusion criteria were as follows: (1) MHs secondary to other causes, such as trauma or high myopia; (2) absence of the ‘cuff’ sign on OCT after primary PPV; and (3) significant dioptric media opacity that impaired OCT and best-corrected visual acuity (BCVA) evaluation.

All patients underwent phacoemulsification cataract extraction, intraocular lens implantation, 25G PPV, ILMP and air tamponade during the initial surgery. Postoperatively, patients were instructed to maintain a prone or face-down position for 3 d. One week after surgery, OCT revealed that the MHs unclosed. The following day, after releasing 0.1 mL of aqueous humour via anterior chamber paracentesis, a pars plana intravitreal injection of 0.3 mL of perfluoropropane (C₃F₈) was administered using a 27G needle under topical anaesthesia. In cases of elevated intraocular pressure (IOP), anterior chamber paracentesis was performed to release a small amount of aqueous humour and normalise the IOP. Patients were again instructed to maintain a prone position for 3 d post-injection.

Comprehensive assessments, including BCVA, IOP measurement, slit-lamp microscopy, indirect ophthalmoscopy, colour fundus photography, and OCT, were conducted before PPV, at 1 week postoperatively, and during follow-up at 1–3 months after intravitreal gas injection. The minimum linear diameter (MLD) of the MH was defined was defined as the linear distance between the two closest points on the MH edge as visualised on horizontal OCT scans [4].

Statistical analysis

Data were analysed using GraphPad Prism 9. Since none of the data were normally distributed, the Wilcoxon signed-rank test was used to evaluate the correlation between MLD and BCVA before PPV, after intravitreal gas injection, and the change in BCVA after intravitreal gas injection. A P-value of < 0.05 was considered statistically significant.

Results

Eight eyes from eight patients (seven females, one male; median age, 65 years, range 62–70 years) were included in the study. General patient data are presented in Table 1. The MLD of the MHs before PPV was 555.4 ± 104.8 μm, which decreased to 335.1 ± 74.3 μm 1 week after PPV (P = 0.008). OCT examination 1 week after PPV showed that the MHs had shrunk, the hole edge was turned up, and a small number of intraretinal cysts were still visible with characteristic “cuff” sign. The macular microstructure of all the patients was improved by OCT examination at 3 months after perfluoropropane tamponade, such as the recovery of intraretinal cysts and continuity of ellipsoid zone.

The mean BCVA was 0.79 ± 0.26 logMAR (20/125) before PPV, and improved slightly (0.68 ± 0.17 logMAR (20/100)), while the MH remained open following primary PPV. Through second operation of intravitreal injection of 0.3 mL of perfluoropropane (C₃F₈), the BCVA improved to 0.48 ± 0.19 logMAR (20/63) 3 months after intravitreal gas injection. This improvement in BCVA was statistically significant (P = 0.01). No significant postoperative complications, such as ocular hypertension or retinal detachment, were observed in any of the patients. Typical cases are illustrated in Figs. 1 and 2.

Table 1 General datas of included patients

Fig. 1

Serial images of case 1 before and after pars plana vitrectomy (PPV) and after intravitreal C₃F₈ injection a Ultra-wide-field fundus image demonstrating a full-thickness macular hole (MH). b–e Serial optical coherence tomography (OCT) images of the case. b Preoperative OCT showing an idiopathic full-thickness MH before PPV. c One week after PPV, OCT revealed a reduction in the MH size and absorption of subretinal fluid. d One week after intravitreal gas injection, OCT indicated MH closure. e During follow-up after the intravitreal gas injection, complete closure of the MHs was achieved

Fig. 2

Serial images of case 2 before and after pars plana vitrectomy (PPV) and after intravitreal C₃F₈ injection. a Colour fundus photograph illustrating a full-thickness macular hole (MH). b–e Serial optical coherence tomography (OCT) image of the case. b Preoperative OCT revealing an idiopathic full-thickness MH with cystoid changes at the edges. c One week after PPV, OCT indicated non-closure of the MH, with minimal intraretinal cysts still present. d One week after intravitreal gas injection, OCT showed closure of the MH, although some disorganisation of the foveal microstructure persisted. e Three months after the intravitreal gas injection, OCT demonstrated that the macular hole was closed and foveal microstructure had improved

Discussion

Despite the high success rate of IMH surgeries, non-closure of MHs remains one of the most common complications following primary PPV [5]. The underlying causes of non-closure in IMHs after primary PPV are poorly understood but might be influenced by factors such as the course of the IMH, MH size, the type of tamponade agent, ILMP, and the extent of peeling. In this study, eight patients underwent phacoemulsification cataract extraction, intraocular lens implantation, 25-gauge vitrectomy, ILMP, and air tamponade as their primary surgeries. An examination on the first postoperative day revealed that only approximately 30–50% of air in the vitreous cavity remained (data not shown). This suggests that air leakage from the incision or inadequate posturing might have reduced the effective tamponade time, which appears to be a primary factor contributing to the non-closure of MHs after the initial operation.

For patients in whom primary surgical interventions fail, there is no consensus regarding the optimal retreatment strategy. Various assistive techniques, including expanding the range of ILMP, using ILM tamponade, performing ILM flap reversal, conducting MH edge massage, making radial incisions in the nerve fibre layer, administering autologous platelets, and using silicone oil tamponade, have been used to treat IMH after primary surgical failure [6]. While these adjunctive methods have improved the surgical success rate, they often necessitate a repeat PPV, leading to additional complexity and increased trauma for the patient.

Hillenkamp et al. [7] used OCT to evaluate the efficacy of reoperation for IMHs after primary PPV failure. They reported that MHs exhibiting a turned-up edge and ‘cuff’ of subretinal fluid were more likely to close during the second operation compared to those without these characteristics. Rishi et al. [8] performed a reoperation involving fluid-air exchange with 14% C₃F₈ gas injection, without expanding the range of ILMP, in a patient who had previously undergone PPV. Complete closure of the MH was observed at the 2-week follow-up after the reoperation. Additionally, Singh et al. [9] reported successful outcomes in three patients with IMHs who had failed primary PPV, using a simple fluid-air exchange followed by 14% C₃F₈ gas injection. These patients also demonstrated closure of the MHs and improvement in visual acuity 6 weeks post-surgery. Notably, the last two clinical case reports indicated that the patients presented with a ‘cuff’ of subretinal fluid before surgery, leading to anatomical closure of the MHs after gas injection. Hillenkamp and Rishi [7, 8] hypothesised that the pathogenesis of MHs closure primarily involves the centripetal movement of retinal tissue. In MHs characterised by the ‘cuff’ sign, there is a lack of adhesion to the retinal pigment epithelium at the edges and beneath the MHs. After gas injection, the retinal tissue moves toward the centre, promoting closure. Conversely, in MHs without the ‘cuff’ sign, closer adhesion to the retinal pigment epithelium might hinder closure, thus reducing the possibility of closure.

Another crucial factor for achieving MHs closure is maintaining a dry environment at the macular interface. This can be achieved by filling the vitreous cavity with gas or air, which prevents fluid from flowing into the subretinal space and restricts the entry of cellular components and growth factors that could hinder healing [10, 11]. Previous studies have indicated that most MHs close by the third postoperative day [12]. C₃F₈ has a longer tamponade duration than air owing to its expandable properties, which provide sustained support for maintaining a dry environment at the macular interface. A volume of 0.3 mL of C₃F₈ is sufficient to cover the MHs, thereby optimising the chances of successful closure.

A previous study indicated that patients with unclosed MHs after primary PPV but exhibiting a smaller or no change in size had a closure rate of up to 80% after reoperation [12]. All patients in that study presented with unclosed MHs after primary PPV, yet the holes were smaller than their preoperative dimensions. OCT analysis of these patients revealed wrapped hole edges accompanied by the characteristic ‘cuff’ sign. Drawing inspiration from these findings, we administered a simple intravitreal injection of 0.3 mL of C₃F₈ while ensuring a strict prone position postoperatively. Remarkably, all patients achieved MHs closure, and visual acuity improved to varying degrees. However, this study has several limitations: (1) the sample size was relatively small and (2) there was a lack of comparative data. Therefore, a prospective study with a larger sample size is required to validate these findings.

In conclusion, for patients with IMHs who experience failure following the first PPV operation, an intravitreal injection of 0.3 mL of C₃F₈ alone presents a straightforward, practical, and effective retreatment option, particularly when OCT examination reveals a ‘cuff’ sign.