Ophthalmic Surgery, Lasers and Imaging Retina

Technique 

A Modified Perfluoro-n-octane-Assisted Autologous Internal Limiting Membrane Transplant for Failed Macular Hole Reintervention: A Case Series

Sengul Ozdek, MD; Prabu Baskaran, MS, DNB; Levent Karabas, MD; Pedro Pereira Neves, MD

Abstract

The authors describe a modified perfluoro-n-octane (PFO)-assisted autologous internal limiting membrane (ILM) transplantation technique for macular hole (MH) reintervention and present results from a series of 11 patients. The authors harvested a free ILM flap and transplanted it into the MH under a PFO bubble. The time at which PFO is injected, the extent of coverage of PFO, and the sequence of fluid-air exchange (FAE) are crucial to overcome previously described technical difficulties of relieving the flap from forceps, stabilizing the flap into the MH, and prevention of flap dislodgement during FAE. A successful U-shaped closure was observed in 10 of 11 cases (90.9%). One case (9.1%) showed flat open closure. The postoperative visual gain was statistically significant (P = .01).

[Ophthalmic Surg Lasers Imaging Retina. 2017;48:416–420.]

Abstract

The authors describe a modified perfluoro-n-octane (PFO)-assisted autologous internal limiting membrane (ILM) transplantation technique for macular hole (MH) reintervention and present results from a series of 11 patients. The authors harvested a free ILM flap and transplanted it into the MH under a PFO bubble. The time at which PFO is injected, the extent of coverage of PFO, and the sequence of fluid-air exchange (FAE) are crucial to overcome previously described technical difficulties of relieving the flap from forceps, stabilizing the flap into the MH, and prevention of flap dislodgement during FAE. A successful U-shaped closure was observed in 10 of 11 cases (90.9%). One case (9.1%) showed flat open closure. The postoperative visual gain was statistically significant (P = .01).

[Ophthalmic Surg Lasers Imaging Retina. 2017;48:416–420.]

Introduction

Though various techniques have been described for closing a large macular hole (MH),1,2,3 options available for failed MH closure are limited. Yuki et al.4 used viscoelastic material to retain the transplanted internal limiting membrane (ILM) flap in the MH but have not clarified if the viscoelastic is completely removed. Also, there is no comment on occurrence of flap dislodgment at the time of fluid-air exchange (FAE). Park et al.5 used a small rolling perfluoro-n-octane (PFO) bubble to relieve the flap from the forceps that holds it over the MH. The rolling is achieved mechanically by rotating the globe. However, there is a potential risk of retinal injury and flap dislodgement during this maneuver. Also, the flap may remain stuck to the forceps, causing difficulty in releasing it.

Flap dislodgement is one of the major factors that determines the success rate of ILM flap transplantation techniques. We report an elegant and reproducible modification of the existing PFO-assisted ILM transplantation technique that simplifies flap handling throughout the procedure without dislodgement at any point during the surgery. This greatly improves the success rate of MH closure in difficult reinterventions.

Patients and Methods

This multicentric, retrospective case study was conducted by four different surgeons (SO, PB, LK, PN), each contributing to this series of 11 patients (Table). The study adhered to the tenets of the Declaration of Helsinki.


            Clinical Characteristics and Surgical Outcome of Clinical Cases

Table:

Clinical Characteristics and Surgical Outcome of Clinical Cases

Surgical Technique

The technique was developed by one of the authors (SO) and adopted by the others. A standard 23- or 25-gauge three-port pars plana entry was followed by injection of Brilliant Blue G (BBG) (0.2 mL of 0.025% BBG) dye over the macular area to visualize the extent of ILM peeled during the first surgery (Figure 1A). PFO was then introduced to cover the macular area, optic disc, and arcade vessels. Next, a small ILM free flap was harvested from the edge of the pre-peeled area using forceps and simply dragged into the MH under the PFO bubble. If the pre-peeled area is small, the free end of the harvested flap can easily be placed within the MH without disturbing the other end that functions like a hinge (hinge flap) (Figures 1B and 1C). The hinge area can then be peeled forming a completely free flap and placed into the MH (Figures 1D and 1E). We place the graft into the crater of the hole in two to three folds. The time at which PFO is injected and the extent of coverage of PFO are crucial as these details simplify all further maneuvers. Next, FAE is carefully done outside the PFO bubble in order to aspirate the fluid completely before approaching the PFO bubble. The chance of dislodgement of flap is very minimal or close to none at the time of FAE owing to the complete aspiration of fluid before PFO. Finally, the PFO is completely removed (Figure 1F) and air-gas exchange is performed by using either SF6 (SO: 20%; PN: 25%) or C3F8 (PB, LK: 10%) (Videos 1 and 2). Face-down position (reading position during day time and avoiding face-up position during sleeping) is recommended for 4 days to 7 days.


            A modified perfluoro-n-octane (PFO)-assisted free flap internal limiting membrane (ILM) transplantation technique (Case 3). (A) Pre-peeled ILM edge following Brilliant Blue G staining. (B) Harvesting an ILM flap under PFO. (C) Placing the free edge of flap over macular hole (MH) under PFO. (D) Converting the flap into a completely free ILM graft by releasing the hinge. (E) Free ILM flap into MH under PFO (dark blue colored). (F) ILM flap in situ after fluid-air exchange.

Figure 1.

A modified perfluoro-n-octane (PFO)-assisted free flap internal limiting membrane (ILM) transplantation technique (Case 3). (A) Pre-peeled ILM edge following Brilliant Blue G staining. (B) Harvesting an ILM flap under PFO. (C) Placing the free edge of flap over macular hole (MH) under PFO. (D) Converting the flap into a completely free ILM graft by releasing the hinge. (E) Free ILM flap into MH under PFO (dark blue colored). (F) ILM flap in situ after fluid-air exchange.

Results

A successful U-shaped closure (Type 1) was observed in 10 of 11 cases (90.9%) (Figure 2). Only one case (9.1%) showed flat open closure (Type 2). Flap dislodgement was not observed in any of our cases intraoperatively. There was no need to reapply PFO or repeat any of the maneuvers to retain the flap in the MH. The mean pre- and postoperative visual acuity were 0.87 logMAR (Snellen 20/160) and 0.41 logMAR (Snellen 20/50), respectively. The visual improvement was statistically significant, with a P value of 0.01 (Wilcoxon signed ranks test). The MHs in this series had a mean minimum diameter of 512 μm ± 208 μm (range: 219 μm to 752 μm) and a mean maximum diameter of 1,088 μm ± 456 μm (range: 379 μm to 1729 μm). The mean follow-up period was 7.64 months ± 5.42 months (range: 2 months to 16 months) (Table). The statistical analysis was done using SPSS software, version 20 (IBM, Armonk, NY).


            Complicated macular hole (MH) secondary to submacular hemorrhage in a 19-year-old woman with acute myeloid leukemia (Case 2). Patient could only be operated on 2 years after presentation because of thrombocytopenia and general health issues pertaining to her systemic illness. (A) At presentation: Optical coherence tomography (OCT) showed atrophic edges of MH without any cystoid changes and atrophic temporal outer retinal layers. (B) After primary surgery: OCT showed persistent MH. (C) After free flap surgery (immediate post op): OCT showed closed MH with internal limiting membrane graft. (D) Four months later: OCT showed MH closure with atrophic outer retinal layers. Final best-corrected visual acuity (BCVA) improved to 1.17 logMAR (Snellen 20/320) from her presenting BCVA of 1.8 logMAR (Snellen 20/1,200).

Figure 2.

Complicated macular hole (MH) secondary to submacular hemorrhage in a 19-year-old woman with acute myeloid leukemia (Case 2). Patient could only be operated on 2 years after presentation because of thrombocytopenia and general health issues pertaining to her systemic illness. (A) At presentation: Optical coherence tomography (OCT) showed atrophic edges of MH without any cystoid changes and atrophic temporal outer retinal layers. (B) After primary surgery: OCT showed persistent MH. (C) After free flap surgery (immediate post op): OCT showed closed MH with internal limiting membrane graft. (D) Four months later: OCT showed MH closure with atrophic outer retinal layers. Final best-corrected visual acuity (BCVA) improved to 1.17 logMAR (Snellen 20/320) from her presenting BCVA of 1.8 logMAR (Snellen 20/1,200).

Discussion

There are various techniques described in the literature for closing a large MH — namely, arcuate retinotomy, massaging the edges, broad ILM peeling from arcade to arcade, inverted ILM flap technique, etc.1,2,3 However, there are only very limited options available for a failed MH, where the ILM has already been peeled during the first surgery. Yuki et al. introduced the concept of autologous ILM transplantation.4 The rationale behind keeping the ILM flap into the MH was to promote gliosis from inside the retina in addition to edges.1 The ILM flap has Müller cell fragments that are known to play a role in glial tissue proliferation.

Whether it is an inverted flap technique, as described by Michalewska et al.,1 or a free flap technique, as described by Yuki et al.,4 we believe that retaining the flap at the time of FAE is the major factor that determines the success of the procedure. In fact, the major limitation stated by Michalewska et al. in their original article on inverted flap technique was dislodgement of the flap (seven out of 50 eyes; 14%) at the time of FAE.1 In a case series of 10 patients, Yuki et al. used viscoelastics to retain the flap over the MH.4 However, there is no mention of flap dislodgement. Also, it is not clear if viscoelastics were completely removed at the time of FAE. Complete removal can result in flap dislodgement, and incomplete removal can have the risk of intraocular pressure elevation.

Park et al. reported a case series of three patients in whom the authors performed PFO-assisted ILM transplantation.5 However their technique is tedious and used a much smaller PFO bubble to stabilize the flap over the MH. Hence, it requires eyeball rotation to make the PFO bubble roll over the flap, before releasing the flap from the forceps, thus potentiating the risk of retinal injury by forceps. There can also be a tendency for the ILM flap to get stuck to the forceps precluding easy release. Additionally, the smaller PFO bubble can move away due to fluid turbulence during exchange of instruments, especially in a nonvalved system, resulting in flap dislodgement.

In our technique of PFO-assisted ILM transplantation, a single large PFO bubble covering the entire posterior pole is injected right at the beginning of the surgery. All steps — including flap harvest, flap positioning into MH, and relieving the flap from forceps — are done under PFO. This makes all maneuvers simple without any need for mechanically rotating the eyeball. Additionally, there is no risk of retinal injury or flap dislodgement intraoperatively. Our technique does not require any other additional maneuvers such as centripetal pulling or massaging of MH edge. Furthermore, in our modification, the flap does not change its medium unlike other techniques (fluid to PFO5 or fluid to viscoelastics4). This is probably why sticking of the flap to the forceps does not occur in our technique.

The possible concern of PFO entrapment within the MH can be addressed by injecting a single large PFO bubble and by preventing formation of fish eggs. We routinely advise our patients to adopt reading position during daytime and to avoid face up position during the night. Proper positioning is started immediately after surgery. For patients who cannot maintain the recommended positioning, we advise to avoid face-up position at least during daytime. We believe that the initial few hours are very crucial for the flap to remain in situ. The nonclosure of MH in one of our patients could be due to flap dislodgement during the immediate postoperative period, probably a result of poor positioning. The only concern with this technique is that it requires an additional consumable (PFO), thereby increasing the cost of surgery marginally.

Conclusion

To conclude, we recommend our modification of PFO-assisted ILM transplantation technique as a safe and effective surgical option for failed MH closures. The technique is very promising in retaining the flap during the crucial step of FAE, resulting in favorable outcomes. The major limitations of our study are small sample size and retrospective design. A prospective, randomized, controlled trial is needed to further assess visual improvement and restoration of foveal microstructures in the future.

References

  1. Michalewska Z, Michalewski J, Adelman RA, Nawrocki J. Inverted internal limiting membrane flap technique for large macular holes. Ophthalmology. 2010;117(10):2018–2025. doi:10.1016/j.ophtha.2010.02.011 [CrossRef]
  2. Charles S, Randolph JC, Neekhra A, Charles D, Littlejohn N, Calzada JI. Arcuate retinotomy for the repair of large macular holes. Ophthalmic Surg Lasers Imaging Retina. 2013;44(1):69–72. doi:10.3928/23258160-20121221-15 [CrossRef]
  3. Iezzi R, Kapoor KG. No face-down positioning and broad internal limiting membrane peeling in the surgical repair of idiopathic macular holes. Ophthalmology. 2013;120(10):1998–2003. doi:10.1016/j.ophtha.2013.06.001 [CrossRef]
  4. Yuki M, Shiraga F, Kimura S, et al. Autologous transplantation of the internal limiting membrane for refractory macular holes. Am J Ophthalmol. 2014;157(4):861–869. doi:10.1016/j.ajo.2013.12.028 [CrossRef]
  5. Park SW, Pak KY, Park KH, Kim KH, Byon IS, Lee JE. Perfluoro-n-octane assisted free internal limiting membrane flap technique for recurrent macular hole. Retina. 2015;35(12):2652–2656. doi:10.1097/IAE.0000000000000754 [CrossRef]

Clinical Characteristics and Surgical Outcome of Clinical Cases

Patient No. Age Sex Duration of Failed MH (Months) Minimal Diameter of MH (µ) Maximum Diameter of MH (µ) Maximum Height of MH (µ) Gas Tamponade MH Closure Flap Dislodgement Preoperative BCVA (logMAR) Postoperative BCVA (logMAR) BCVA at Last Visit (logMAR) Follow-Up (Months)
1 65 M 12 676 1374 612 SF6 20% Yes No 1.3 0.8 0.4 14
2 19 F 6 707 967 353 SF6 20% Yes No 1.8 1.3 1.17 2
3 67 F 6 255 576 543 SF6 20% Yes No 0.7 0.7 0.4 1.5
4 62 F 3 219 379 275 C3F8 10% No No 1.48 2 0.3 9
5 63 M 9 678 1708 635 C3F8 10% Yes No 0.5 2 0.3 16
6 75 F 1 410 1040 555 C3F8 10% Yes No 0.2 2 0.3 13
7 72 F 2 705 1729 636 C3F8 10% Yes No 1.3 2 0.15 12
8 60 M 1 752 1209 378 C3F8 10% Yes No 1 0.6 0.3 7
9 53 M 1 584 1500 548 C3F8 10% Yes No 0.8 0.8 0.60 4
10 57 F 1 256 808 486 C3F8 10% Yes No 0.8 0.3 0.3 1
11 67 F 9 390 685 389 SF6 25% Yes No 0.8 0.3 0.3 4
Authors

From Gazi University, School of Medicine, Besevler, Ankara, Turkey (SO); Aravind Eye Hospital, Pondicherry, India (PB); Kocaeli University, School of Medicine, Umuttepe, Kocaeli, Turkey (LK); and Setúbal Hospital Center, Setúbal, Portugal (PPN).

The authors report no relevant financial disclosures.

Address correspondence to Prabu Baskaran MS, DNB, Aravind Eye Hospital, Cuddalore Main Road, Pondicherry – 605007, India; email: prabubaskaran@gmail.com.

Received: December 10, 2016
Accepted: March 01, 2017

10.3928/23258160-20170428-08

Sign up to receive

Journal E-contents