In vivo imaging of the outer retina after an autologous neurosensory retinal free flap transplantation for a refractory macular hole closure

Abstract

Background

There are limited surgical options to successfully close a refractory macular hole. One promising option is an autologous neurosensory retinal free flap transplantation. An autologous neurosensory retinal free flap transplantation places a graft of peripheral autologous retinal tissue into the macular hole and was developed to improve post-surgical outcomes. Here, clinical instrumentation and a high-resolution adaptive optics system imaged the graft and host tissue of a patient whose refractory macular hole was successfully closed with an autologous neurosensory retinal free flap transplantation.

Case presentation

A 71-year-old Hispanic female with bilateral moderate nonproliferative diabetic retinopathy (visual acuity of 20/100 in each eye) underwent an autologous neurosensory retinal free flap transplantation in the right eye only, which successfully closed a large refractory macular hole measuring 4° in diameter. Although somewhat variable, the best-corrected visual acuity improved from 20/100 to 20/70 with a subjective improvement noted by the patient. The eye was examined using (1) fundus photography and (2) clinical optical coherence tomography both presurgery and post surgery and (3) with adaptive optics–optical coherence tomography–scanning laser ophthalmoscopy post surgery. Postsurgical clinical optical coherence tomography imaging revealed restoration of the external limiting membrane within the graft. Adaptive optics–optical coherence tomography imaging provided enhanced lateral and axial resolution and showed a restored inner segment/outer segment junction within the graft. Adaptive optics–optical coherence tomography also revealed the cone outer segment tip layer in the host tissue, highlighting preservation of the microarchitecture and indicating that the host tissue was not negatively impacted by the surgery or the presence of the graft. Further, adaptive optics–scanning laser ophthalmoscopy imaging revealed photoreceptors within the graft and surrounding host tissue, indicating surgical success, graft acceptance and viable host tissue.

Conclusion

Although the exact physiological mechanisms that promote macular hole closure and intraretinal cellular changes after an autologous neurosensory retinal free flap transplantation are unknown, imaging supports the procedure as a reasonable surgical option for refractory macular hole closure. The preserved integrity of the host tissue suggests that the graft does not negatively impact the retina following the surgery. Furthermore, the improvement in the inner segment/outer segment junction and external limiting membrane noted over time within the graft are considered favorable as they relate to the structure and function of the retina.

Background

Surgery for the repair of full thickness macular holes (MH) is generally successful, with closure rates exceeding 90% [1, 2]. However, the closure rates for refractory MHs are lower, with limited improvements in the final best-corrected visual acuity (BCVA) upon closure [3]. In fact, once the posterior hyaloid has been removed and the internal limiting membrane (ILM) has been peeled, there are limited surgical options to address a refractory MH. An autologous neurosensory retinal free flap transplantation (ANRFFT) is a procedure that places a graft of peripheral autologous retinal tissue into the MH and was developed to improve postsurgical outcomes [4]. Here, a large refractory MH measuring 4° in diameter was closed with an ANRFFT and was subsequently imaged with an adaptive optics–optical coherence tomography–scanning laser ophthalmoscopy (AO–OCT–SLO) system to further investigate the outer retinal microstructure within and surrounding the graft.

Case presentation

A 71-year-old Hispanic female with bilateral moderate nonproliferative diabetic retinopathy underwent an ANRFFT in the right eye only, which had a large refractory MH measuring 4° in diameter and a BCVA of 20/100 (Fig. 1a, b). Two previous surgeries to address the MH had been attempted, including a pars plana vitrectomy (PPV) and an ILM peel with a flap, and both had failed to close the hole. Given the limited surgical options following the detachment of the posterior hyaloid and removal of the ILM, the decision was made to perform an ANRFFT. A full thickness neurosensory retinal free flap was harvested from the superior retina, ~20° from the fovea (Fig. 1c). The flap was moved to the MH under perfluoro-n-octane, the edges were flattened and tucked into the MH. Silicone oil was used as the intraocular tamponade and was removed in a subsequent procedure. The patient was imaged prior to surgery (Fig. 1a, b), at 5 months post surgery (Fig. 1c, d), at 10 months post surgery (Fig. 1e, f), and at 26 months post surgery (Fig. 1g, h) with fundus photography (P200DTX; Optos, Dunfermline, UK) and clinical OCT (Cirrus HD-OCT 5000; Zeiss, Oberkochen, Germany). The graft was noted to be contiguous with the host tissue within 2 weeks of surgery by clinical OCT, and the BCVA was measured at 20/100. Structural changes continued throughout the postoperative period. Inner-retinal cysts that were noted to increase in size and number and improved with steroid therapy, suggestive of an inflammatory component (Fig. 1d, f). At 10 months post surgery, the BVCA improved to 20/70. At the most recent clinical follow-up, 26 months post surgery, irregularities in the contour of the inner retina had smoothed out, and the BCVA remained constant at 20/70. However, a few of the cysts did persist, which also indicate structural changes (Fig. 1h).

Fig. 1.

Fundus photographs and clinical optical coherence tomography B-scans from the right eye at four time points. The B-scans are 20° scans across the macula. The dashed white line on each of the fundus photographs indicates the location of the clinical optical coherence tomography B-scan. a Fundus photograph prior to surgery with the macular hole visible. b Clinical optical coherence tomography B-scan prior to surgery reveals a full thickness macular hole measuring 4° in diameter. c Fundus photograph of the closed macular hole and graft origination site at 5 months post surgery. d Clinical optical coherence tomography B-scan of the macula at 5 months post surgery with inner-retinal cysts. e Fundus photograph and graft origination site at 10 months post surgery. f Clinical optical coherence tomography B-scan of the macula at 10 months post surgery. g Fundus photograph and graft origination site at 26 months post surgery. h Clinical optical coherence tomography B-scan of the macula at 26 months post surgery with a marked decreased in the inner-retinal cyst severity. c, e, g Blue arrows indicate graft origination site. d, f, h Orange arrows indicate the edges of the graft

An AO–OCT–SLO [5] was used to image the outer retinal layers at four retinal locations in the macula 6° temporal retina (TR) and 6° nasal retina (NR) from the fovea (outside the graft) and at 2° TR and 2° NR (within the graft). All locations were imaged at 5 and 30 months post surgery. The tenets of the Declaration of Helsinki were observed, and the protocol was approved by the Institutional Review Board of The Ohio State University. Written informed consent was obtained after procedures were explained and prior to measurement.

Discussion

Left untreated, a full thickness MH can enlarge or lead to retinal detachment with subsequent vision loss [6]. About 6% will spontaneously close; however, advanced or large holes rarely spontaneously close, and only 15 spontaneous closures of large holes have been documented [7]. PPV was introduced in 1991 as a surgical option to close MHs [8]. Advancements in surgical techniques, including PPVs, have led to reports of 95% success rates [9]. However, the success rate is lower for holes that may be larger or chronic, with reports suggesting closure rates of 69% or lower, and BCVA is only moderately improved when compared with observation alone [10]. An ANRFFT was reported in 2016 as a surgical option when traditional surgical approaches do not successfully close a MH [4]. Studies report ANRFFT closure rates nearing 90% and BCVA improvements up to five Snellen lines, although results can be variable [11–13]. For instance, mean pre- to postoperative BCVA range from 20/258 to 20/214, from 20/362 to 20/231, and from 20/678 to 20/155 [11–13]. Furthermore, at follow-up, restoration of the outer retinal layers including the ellipsoid zone and external limiting membrane (ELM) have been reported in a majority of eyes [11, 12]. The physiological mechanisms behind the closure and restoration of the outer layers are still an active area of research, but it is thought that the graft acts as a scaffold for the proliferation of Müller cells that secrete neurotrophic and growth factors, which can increase retinal neural cell survival as well support migration of cells from surrounding tissue [11, 13, 14]. As with any surgery, there are risks such as retinal detachment, vitreous hemorrhage, cystoid macular edema, reactive pigment epithelia hyperplasia, and sizing errors of the graft within the MH [11–13]. These complications have been reported in ~8.5% of eyes [12].

In this case, the preoperative BCVA was 20/100, better than then mean pre- and postoperative BCVA reported in literature [11–13]. Given the patient’s BCVA was relatively good to begin with, it is understandable that the improvement was less than the 5 Snellen lines previously reported [11–13]. The postoperative BCVA, 20/70, is ultimately an excellent result for a refractory MH closure with an ANRFFT.

Here, the clinical OCT imaging, Fig. 1, shows findings that are consistent with previous reports in literature [11–13]. Figure 2 shows the clinical OCT images at 5 and 30 months post surgery with the location of the AO–OCT–SLO imaging marked with the colored arrows and boxes. Figure 2a is the same clinical OCT image shown in Fig. 1d, and Fig. 2b is the same clinical OCT image shown in Fig. 1h. Post surgery, the host tissue at 6° TR remained stable, and the retinal pigment epithelium (RPE) was present, but the ELM and IS/OS junctions were not fully intact or completely resolved (Fig. 2a, b). At 6° NR, no significant disruptions of microarchitectural OCT structure in the outer retina of host tissue were observed either presurgery or post surgery; clearly resolved and continuous ELM, IS/OS junction, and RPE were present (Fig. 2a, b). Throughout the observation period, the graft tissue showed a restoration of the ELM. Initially at 5 months, there was a faint and discontinuous ELM, Fig. 2a. At 26 months, the ELM was brighter and more continuous, as indicated by the yellow arrow in Fig. 2b.

Fig. 2.

Adaptive optics–optical coherence tomography–scanning laser ophthalmoscopy images and clinical optical coherence tomography images taken at 5 and 30 months post surgery. a Clinical optical coherence tomography B-scan of the macula at 5 months post surgery, from Fig. 1d, with the four adaptive optics retinal imaging locations labeled; the yellow dot corresponds to the center of the graft. b Corresponding clinical optical coherence tomography B-scan of the macula at 26 months post surgery, from Fig. 1h, with the four adaptive optics retinal imaging locations labeled. c Adaptive optics–optical coherence tomography from 6° temporal retina taken 5 months post surgery. d Adaptive optics–optical coherence tomography from 6° temporal retina taken 30 months post surgery. e Adaptive optics–optical coherence tomography from 2° temporal retina taken 5 months post surgery. f Adaptive optics–optical coherence tomography from 2° temporal retina taken 30 months post surgery. g Adaptive optics–optical coherence tomography from 2° nasal retina taken 5 months post surgery. h Adaptive optics–optical coherence tomography from 2° nasal retina taken 30 months post surgery. i Adaptive optics–optical coherence tomography from 6° nasal retina taken 5 months post surgery. j Adaptive optics–optical coherence tomography from 6° nasal retina taken 30 months post surgery. k Adaptive optics–optical coherence tomography from 6° temporal retina taken 5 months post surgery. l Adaptive optics–optical coherence tomography from 2° temporal retina taken 5 months post surgery. m Adaptive optics–scanning laser ophthalmoscopy from 2° nasal retina taken 5 months post surgery. n Adaptive optics–scanning laser ophthalmoscopy from 6° nasal retina taken 5 months post surgery. c–j Cyan arrows indicates the IS/OS junction, yellow arrows indicate the external limiting membrane, and pink arrows indicate the cone outer segment tip layer. k, n Green arrows indicate photoreceptors. Scale bar is 50 µm

AO–OCT provides higher resolution than clinical OCT particularly in the transverse direction; providing enhanced details of the retinal structure. Similar to the clinical OCT, the ELM (yellow arrow) and IS/OS junction (cyan arrow) at 6° TR were not fully intact nor completely resolved (Fig. 2c, d). AO–OCT imaging revealed the cone outer segment tip (COST) layer at 6° NR throughout the observation period, as shown by the bright red arrows in Fig. 2i, j, indicating no negative impact to the host tissue by the graft procedure. On the contrary, the COST was not resolved in the clinical OCT throughout the observation period (Fig. 2a, b). At 5 months, AO–OCT resolved the IS/OS junction, which was intermittently present within the graft at 2° TR and 2° NR, as shown by the cyan arrows in Fig. 2e, g. The IS/OS junctions at 2° TR and 2° NR were not visible on the clinical OCT (Fig. 2a). In agreement with clinical OCT imaging, the ELM was faint and discontinuous at 2° TR and 2° NR at 5 months, as shown by the yellow arrows in Fig. 2e, g). Similar to findings at 5 months, the IS/OS junction was not clearly resolved on clinical OCT at 30 months (Fig. 2b); however, a slightly more distinct and continuous IS/OS junction was observed on AO–OCT at 2° TR and 2° NR, as shown by the cyan arrow in Fig. 2f, h. The ELM remained faint and discontinuous at 2° TR and 2° NR, as shown by the yellow arrows in Fig. 2f, h. These findings indicate continued graft acceptance throughout the observation period and possible tissue restoration.

AO–SLO imaging revealed cone photoreceptors (reflective dots) at 6° NR (Fig. 2k, green arrow) and potentially at 6° TR—Fig. 2n, green arrow—however the ELM and IS/OS junction was not preserved as shown in AO–OCT images throughout this retinal location. The AO–SLO image at 2° TR, (Fig. 2l) is more interesting, and the bright white dots are indicative of photoreceptors, but one would expect a rod-dominated retina at 20° [15]. Such structures would be hard to resolve given the small imaging exit pupil diameter (~5.5 mm) available for this particular subject. The visible ELM and IS/OS junction layers visible in the corresponding AO–OCT images provide evidence that these could be photoreceptors, but whether these are cones or abnormal rods is unknown. For 2° NR, the overlying cyst reduced the AO–SLO imaging quality (Fig. 2m).

Further functional tests such as multifocal electroretinopgraphy, visual field sensitivity or microperimetry, and fluorescein angiography (FA) were not performed during her postoperative care. These additional tests would give further insights into the success of the ANRFFT procedure. Furthermore, FA would provide evidence of successful graft integration by revealing the reperfusion of the superficial and inner retinal vessels as previously reported in literature [16].

Conclusion

To our knowledge, this is the first report to image the graft after an ANFRFFT with AO. Although the exact physiological mechanisms that promote MH closure and tissue repair after an ANFRFFT are unknown [11], the imaging and functional improvement demonstrate, surgical success, and graft acceptance, as well as support the possibility of outer retinal tissue restoration. The preserved integrity of the host tissue suggests that the graft did not negatively impact the retina. As ANRFFTs become more common, further research and in vivo imaging are necessary to elucidate the physiological mechanisms responsible for these observations.

Abbreviations

MH

Macular hole

VA

Visual acuity

BCVA

Best corrected visual acuity

ILM

Internal limiting membrane

ANRFFT

Autologous neurosensory retinal free flap transplantation

AO–OCT–SLO

Adaptive optics–optical coherence tomography–scanning laser ophthalmoscopy

PPV

Pars plana vitrectomy

TR

Temporal retina

NR

Nasal retina

ELM

External limiting membrane

IS/OS

Inner segment/outer segment

COST

Cone outer segment tip

FA

Fluorescein angiography

Author contributions

Data collection, SC and ND; data processing, DJ; original draft preparation, DJ; review, writing and editing final manuscript, DJ, SC, ND and MO; patient referral, MO.

Funding

This work was supported by Department of Defense (DoD) Telemedicine and Advanced Technology Research Center (TATRC) Grant W81XWH-10-1-0738. The supporting sponsor had no involvement in the study design, collection, analysis or interpretation of data, writing the manuscript, or in the decision to submit the manuscript for publication.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The tenets of the Declaration of Helsinki were observed, and the protocol was approved by the Institutional Review Board of The Ohio State University (OSU). Written informed consent was obtained after all procedures were fully explained to the patient and prior to imaging.

Consent for publication

Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.

Competing interests

Danae A. Johnson, Nathan Doble: None, Matthew P. Ohr: Alimera, SH; Apellis I, G; Genentech/Hoffman-LaRoche, I, G; Novartis, I, G; Regeneron, I, G, Vitranu, SH, AB; Stacey Choi: None.