The understanding of vitreomacular interface disorders has improved considerably with the development of high-resolution spectral-domain optical coherence tomography (SD-OCT), facilitating both the study and the classification of each different clinical entity.1
Although OCT-based diagnostic criteria for full-thickness macular hole (FTMH) and non-FTMH have already been described in the literature, significant controversy still exists regarding their pathogenesis, management, and prognosis.2 Indeed, much morphological heterogeneity can be found while examining various macular hole (MH) subtypes, such as lamellar MH (LMH) and macular pseudohole.
A recent subclassification of LMH proposed its division into two distinct entities: tractional LMH and degenerative LMH.3 The first subtype, characterized by the schitic separation of neurosensory retina between the outer plexiform and outer nuclear layers, often presents an intact ellipsoid layer and is associated with tractional epiretinal membranes (ERMs) and / or vitreomacular traction. The second subtype shows the presence of intraretinal cavitation that might affect all retinal layers and is often associated with nontractional epiretinal proliferation4–6 and a retinal “bump.”
Recently, the use of optical coherence tomography angiography (OCTA) showed that macular holes have an altered pattern in the various retinal capillary plexuses,7,8 suggesting that the microvascular environment might play a role in disease pathogenesis. The rationale of our present study was to use OCTA to investigate whether a microvascular change specifically occurs in the degenerative subtype of LMH and to compare it with the healthy fellow eye and with a control group.
Patients and Methods
In this single-center, observational, cross-sectional study, we used swept-source OCT (SS-OCT) and OCTA to compare the vascular density of the superficial capillary plexus (SCP), deep capillary plexus (DCP), and choriocapillaris (CC) in eyes with degenerative LMH and in healthy fellow eyes. Data were then compared with an age- and sex-matched control group. We certify that every applicable institutional and governmental regulation concerning the ethical use of human volunteers was followed during this study. All patients gave written general informed consent to take part in observational studies approved by the institutional review board of the San Raffaele Scientific Institute. The study was carried out in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki.
All patients fulfilling the inclusion/exclusion criteria were consecutively recruited at the Vitreoretinal Surgery Service of the Ophthalmology Department, San Raffaele Scientific Institute, between January and June 2016.
Inclusion criteria were age of 18 years or older, axial length between 22.5 mm and 24.5 mm (measured with IOLMaster; Zeiss, Jena, Germany), and a diagnosis of unilateral idiopathic degenerative LMH. Diagnosis was obtained using biomicroscopy and confirmed with SD-OCT.
Exclusion criteria for the study eyes were concurrent ocular disease other than macular hole involving the posterior segment (eg, uveitis, retinal vein occlusion, glaucoma, optic neuropathy, diabetic retinopathy, age-related macular degeneration), any previous posterior segment surgical or laser treatment, cataract surgery within the past 6 months, past complicated cataract surgery, and any previous intravitreal injection of anti-vascular endothelial growth factor agents or corticosteroids. Eyes with optical media opacities that could interfere with good quality imaging acquisition were also excluded. Patients with a history of arterial hypertension, diabetes mellitus, systemic vasculopathies (eg, vasculitides), or connective-tissue disease were also excluded from the study.
Definition of the degenerative LMH subtype was based on the criteria published by Govetto et al.3: presence of a foveal bump, presence of lamellar hole-associated proliferation, and, in the large majority of cases, a disrupted ellipsoid zone. This subtype of MH was characterized by a round-edged intraretinal cavitation potentially involving the outer retinal layers, rather than a split between the inner and outer retina.
Patients underwent complete ophthalmic examination, including best-corrected visual acuity (BCVA) on Early Treatment Diabetic Retinopathy Study charts, anterior segment biomicroscopy, applanation tonometry, indirect fundus examination, and SS-OCT and OCTA scans of the macula.
The fellow eye was examined in each patient and was enrolled as a control (“unaffected fellow”) in the presence of axial length of 24 ± 1 mm, adherence to the above-mentioned exclusion criteria, and absence of any vitreoretinal interface alteration (eg, ERM, MH, epiretinal proliferation, vitreomacular traction).
We also enrolled a group of healthy subjects as controls. Inclusion criteria were a silent ophthalmic pathological history, BCVA of 20/20 or better, axial length of 24 mm ± 1 mm, normal optic nerve with no neuroretinal rim alterations, anterior chamber with open angle, normal fundus biomicroscopy examination, SD-OCT scan within normal limits, and no previous surgery other than uncomplicated cataract extraction and intraocular lens (IOL) implantation.
We obtained SS-OCT and OCTA scans with DRI OCT Triton (Topcon Corporation, Tokyo, Japan). Macular OCTA was acquired in the 3 mm × 3 mm macular area. Each examination was carried out by a single experienced operator (MG). We used only automatically segmented images of the SCP, DCP, and CC to prevent any subjective influence during the final analysis.
All 3 × 3 mm OCTA images were exported from the OCT database as a Joint Photographic Experts Group (JPEG) file and then transferred to ImageJ 1.48 software (National Institutes of Health, Bethesda, MD) for calculation purposes.
To calculate the vessel density coefficient, images were binarized through a threshold strategy as in previous studies.9–11 We generated a novel macro that automatically (1) converts the image from 8-bit into red, green, and blue (RGB) color type; (2) splits it into the three channels (red, green, and blue), keeping the red one open as a reference; (3) applies a mean threshold to convert the image from gray-scale to binary; (4) converts the processed images back to RGB; and (5) restores the FAZ area and colors it with pure blue. White pixels were considered as vessels, black pixels as the background, and blue pixels were automatically excluded from the analysis (Figure 1). Vessel density was expressed as the ratio between measured vessel pixels and the total 3 × 3 mm area after subtracting the FAZ area. We used this method to analyze the SCP, DCP, and CC of patients and healthy controls.
Binarization example of a lamellar macular hole optical coherence tomography angiography 3 mm × 3 mm scan in the superficial capillary plexus (A). In the binarized picture (B), white pixels represent vascular tissue, whereas black pixels are the remaining non-vascular retinal tissue.
To make a thorough analysis of the capillary plexus density in the immediate proximity of the LMH, without considering alteration due to the FAZ area, we arbitrarily designed a method to investigate the vasculature in the area immediately surrounding the FAZ. A 1.5-mm diameter circle centered in the FAZ was superimposed onto the OCTA image, and the FAZ area was removed before vessel density calculation. This allowed us to perform two different analyses: firstly on a “ring” proximal to the FAZ (named peri-FAZ) and secondly outside the circle (outer analysis) (Figure 2).
The retinal area in the immediate proximity of the foveal avascular zone (FAZ), named peri-FAZ, was outlined as a fovea-centered 1.5-mm diameter ring, after excluding the FAZ (A). The outer retinal region (B) is the area included in the 3 mm × 3 mm scan outside the peri-FAZ. Blue regions are automatically excluded from the vascular density calculation by the software algorithm in each of the two areas.
The FAZ area in the SCP was manually outlined with a polygon selection tool, and its area (in mm2) was calculated using a previously described method.11 The FAZ area was never recognizable in the DCP and was not included in our analysis.
The integrity of the outer layers (ellipsoid zone [EZ] and external limiting membrane [ELM]) was also evaluated, based on the continuity of hyperreflective lines corresponding to EZ and ELM in the foveal area.
Variables included in the analysis were age, sex, eye (right / left), BCVA, FAZ area in the SCP, and vessel density. All data are presented as mean ± standard deviation. All variables were tested for normal distribution using the D'Agostino-Pearson test. Differences between groups for age, BCVA, FAZ area, and vessel density in every layer were assessed with one-way analysis of variance (ANOVA), and Bonferroni correction was used as a post hoc test. We adjusted P values for multiplicity with both the Bonferroni and Holm procedures. Differences between sex and qualitative analyses of OCTA scans were assessed using the chi-square test. Statistical analysis was carried out with GraphPad Prism software 6.0 (GraphPad Software, San Diego, CA). All tests were two-sided; P values < .05 were considered significant.
We enrolled 34 study eyes of 34 patients that fulfilled the inclusion / exclusion criteria. Two patients were excluded, one for optical media opacities and one due to impaired glucose tolerance pending investigation. Hence, we finally considered 32 study eyes and 32 healthy fellow eyes of 32 patients (14 males and 18 females, aged 60.2 years ± 3.4 years). We also included 30 age- and sex-matched control eyes of 30 healthy subjects (15 males and 15 females, aged 59.6 years ± 1.4 years). The age (P = .899), sex (P = .799), and eye laterality (right / left) (P = .999) of both subjects and controls were similar. In the LMH group, BCVA was inferior (0.38 logMAR) compared with unaffected fellow eyes and healthy eyes (P < .001).
All continuous studied variables showed normal distribution. The average time from LMH diagnosis to referral was 3.5 ± 1.4 months.
Vessel density in the peri-FAZ varied greatly among the three subgroups in the SCP (P = .003). In the post hoc analysis, the SCP density of both the LMH (0.36 ± 0.02) and fellow eyes (0.36 ± 0.03) subgroups showed higher vascular density than controls (0.34 ± 0.02; P = .004 for LMH; P = .015 for fellow eyes), but no difference was evident between LMH and fellow eyes (P = .190) (Figure 3). Bonferroni and Holm adjustments maintained the same statistical significance: 0.004 adjusted to 0.012, 0.015 to 0.030, and 0.190 kept to 0.190. No significant differences were evident among the three subgroups for DCP vessel density (0.37 ± 0.03 for LMH, 0.36 ± 0.03 for fellow eyes, 0.37 ± 0.03 for controls; P = .231). Similar findings were evident in CC density (0.51 ± 0.04 for LMH, 0.52 ± 0.03 for fellow eyes, 0.51 ± 0.03 for controls; P = .658).
Box and whisker plots showing the retinal capillary density of the superficial (A) and deep (B) capillary plexus in degenerative lamellar macular holes and unaffected fellow eyes compared with healthy controls. The capillary density was calculated in the immediate proximity of the foveal avascular zone (FAZ), named peri-FAZ, outlined as a fovea-centered 1.5-mm diameter ring, after excluding the FAZ.
Vessel density in the area outside the peri-FAZ (outer analysis) did not vary among the three groups in the SCP (0.45 ± 0.03 for LMH, 0.45 ± 0.02 for fellow eyes, 0.46 ± 0.03 for controls; P = .635), in the DCP (0.47 ± 0.04 for LMH, 0.45 ± 0.03 for fellow eyes, 0.45 ± 0.03 for controls; P = .099), or in the CC (0.51 ± 0.04 for LMH, 0.51 ± 0.03 for fellow eyes, 0.51 ± 0.03 for controls; P = .356).
The FAZ area in the SCP turned out to be statistically larger in the LMH (0.39 ± 0.16 mm2) and in the fellow eye (0.39 ± 0.21 mm2) groups compared with controls (0.27 ± 0.07 mm2; P = .021 for LMH; P = .0043 for fellow eyes); otherwise, no difference was present between LMH and fellow eye groups (P = .967). The FAZ area was not recognizable in the DCP of LMH and was not included in our analysis. No difference was evident between fellow eyes (0.67 ± 0.38 mm2) and healthy controls (0.49 ± 0.12 mm2; P = .169).
Structural SD-OCT changes were found only in the LMH group. Fellow eyes and controls, as per protocol, were clinically healthy and did not present any structural OCT alteration. All the LMH eyes presented a round-edged intraretinal cavitation, variably involving the outer retinal layers. Twenty-eight eyes (87.5%) presented lamellar hole-associated epiretinal proliferation (LHEP), whereas 18 (56.2%) presented a foveal bump. We found four study eyes (12.5%) with a loss of integrity of the outer retinal layers, specifically with a change in both the ELM and EZ. An isolated change of a single layer (ELM without EZ or vice versa) was never found.
The pathogenesis of LMH is still under investigation. Various pathways have been proposed, ranging from tractional to nontractional mechanisms, including LHEP, whose role is not yet clear.3–5,12–15 Recent clinical reports show the greater severity and the worse surgical outcome of LMH associated with LHEP compared with those without ERM or with conventional ERM, supporting nontractional pathogenesis.16–18 Furthermore, the reason why some LMH subtypes develop intraretinal splitting while others do not is as yet unknown. One hypothesis suggests a common early event (posterior vitreous detachment with schisis of the hyaloid), followed by divergence into two distinct phenotypes. Another possible explanation supports the theory that the two entities represent different lesions, arising from a distinct pathogenic mechanism. Therefore, the current diagnosis of LMH may actually include more than one condition.
Considering the OCTA alterations described in association with different types of MHs,7,8 one could hypothesize that the degenerative subtype of LMH might be associated with a specific change in the retinal microvascular plexuses. The purpose of our study was to test whether any vascular alteration was evident by specifically analyzing only the degenerative LMH subtype using OCTA. We used a novel strategy to analyze the retinal vascular plexuses in the immediate proximity of the macula, dividing the tested areas in two parts: the “peri-FAZ” and the retinal area outside the peri-FAZ. The peri-FAZ does not include the FAZ but represents a “ring” of tissue around the FAZ. Because our vascular density calculation provides a pure coefficient, representing vessel quantity (white pixels) over the remaining retinal tissue quantity (black pixels), this strategy is independent of the FAZ area dimension. Considering that significant FAZ alteration is always present in LMH,8 we decided to exclude the FAZ from our study to avoid any possible influence. We only focused on the density of the vascular plexuses in the immediate proximity of the macula, which is considered the area most affected by macular disease.
We found significant involvement of the superficial capillary plexus in LMH, it being the only plexus showing increased density compared with healthy eyes. We found this change in the peri-FAZ only, as the retina distant from the FAZ (outer analysis) was not involved in any significant alteration. We speculate that the anatomical cavitated shape of the LMH might push toward the edge of the hole and consequently increase tissue tension. The LHEP, together with the putative vitreous trapped inside, despite the absence of a clear tractional component, might engender the mild tautness responsible for increased vessel density. Accordingly, we found an enlargement of the FAZ area in the SCP of both LMH and fellow eyes; this might be due to development of the central opening of LMH or to traction from hyaloid or LHEP. Vessels are consequently displaced laterally, causing greater vascular density in the perifoveal area.
A second consideration is the absence of any significant differences compared with the clinically healthy fellow eye, as previously found in macular hole at various stages.8 This result implies that the fellow eyes of patients with degenerative LMH may not be considered healthy eyes, despite no clinical or OCT evidence of alteration. Furthermore, this subtle pathologic OCTA change might be considered a slow chronic degenerative process that starts earlier than the clinical manifestation of the disease. Moreover, the bilaterality of the condition requires special attention in every case of unilateral LMH, particularly in regard to follow-up. A long follow-up study indeed suggests that a higher incidence of bilateral disease could develop over a longer time span.19
Considering the relatively low incidence of EZ and ELM changes (12.5%), we were unable to ascertain whether these alterations might influence the microvascular changes observed. However, it is reasonable to infer that the modified vascular density we found was not influenced as the outer vascular layers remained unchanged.
This study has numerous limitations: the small number of patients, the cross-sectional design, and the lack of follow-up. Moreover, we are unable to ascertain whether the changes we found were attributable to LMH presence or to other unknown causes, especially in the healthy fellow eyes. However, strict enrollment criteria, together with robust statistical inferences. allow us to make this consideration.
Despite these limitations and the speculative disposition of our conclusions, we believe our work opens a new debate on the role of vascularization in MHs. Further aspects, particularly the investigation of postoperative changes following macular hole surgery, should certainly be considered for a correct pathophysiological interpretation and for a comprehensive analysis of this clinical aspect.
- Duker JS, Kaiser PK, Binder S, et al. The International Vitreomacular Traction Study Group classification of vitreomacular adhesion, traction, and macular hole. Ophthalmology. 2013;120(12):2611–2619. doi:10.1016/j.ophtha.2013.07.042 [CrossRef]
- Takahashi H, Kishi S. Tomographic features of a lamellar macular hole formation and a lamellar hole that progressed to a full-thickness macular hole. Am J Ophthalmol. 2000;130(5):677–679. doi:10.1016/S0002-9394(00)00626-7 [CrossRef]
- Govetto A, Dacquay Y, Farajzadeh M, et al. Lamellar macular hole: Two distinct clinical entities?Am J Ophthalmol. 2016;164:99–109. doi:10.1016/j.ajo.2016.02.008 [CrossRef]
- Pang CE, Spaide RF, Freund KB. Comparing functional and morphologic characteristics of lamellar macular holes with and without lamellar hole-associated epiretinal proliferation. Retina. 2015;35(4):720–726. doi:10.1097/IAE.0000000000000390 [CrossRef]
- Pang CE, Spaide RF, Freund KB. Epiretinal proliferation seen in association with lamellar macular holes: A distinct clinical entity. Retina. 2014;34(8):1513–1523. doi:10.1097/IAE.0000000000000163 [CrossRef]
- Itoh Y, Levison AL, Kaiser PK, Srivastava SK, Singh RP, Ehlers JP. Prevalence and characteristics of hyporeflective preretinal tissue in vitreomacular interface disorders. Br J Ophthalmol. 2016;100(3):399–404. doi:10.1136/bjophthalmol-2015-306986 [CrossRef]
- Pierro L, Iuliano L, Bandello F. OCT angiography features of a case of bilateral full-thickness macular hole at different stages. Ophthalmic Surg Lasers Imaging Retina. 2016;47(4):388–389. doi:10.3928/23258160-20160324-16 [CrossRef]
- Pierro L, Rabiolo A, Iuliano L, Gagliardi M, Panico D, Bandello F. Vascular density of retinal capillary plexuses in different subtypes of macular hole. Ophthalmic Surg Lasers Imaging Retina. 2017;48(8):648–654. doi:10.3928/23258160-20170802-07 [CrossRef]
- Nemiroff J, Kuehlewein L, Rahimy E, et al. Assessing deep retinal capillary ischemia in paracentral acute middle maculopathy by optical coherence tomography angiography. Am J Ophthalmol. 2016;162:121–132.e1. doi:10.1016/j.ajo.2015.10.026 [CrossRef]
- Chidambara L, Gadde SG, Yadav NK, et al. Characteristics and quantification of vascular changes in macular telangiectasia type 2 on optical coherence tomography angiography. Br J Ophthalmol. 2016;100(11):1482–1488. doi:10.1136/bjophthalmol-2015-307941 [CrossRef]
- Samara WA, Say EA, Khoo CT, et al. Correlation of foveal avascular zone size with foveal morphology in normal eyes using optical coherence tomography angiography. Retina. 2015;35(11):2188–2195. doi:10.1097/IAE.0000000000000847 [CrossRef]
- Compera D, Entchev E, Haritoglou C, et al. Lamellar hole-associated epiretinal proliferation in comparison to epiretinal membranes of macular pseudoholes. Am J Ophthalmol. 2015;160(2):373–384.e1. doi:10.1016/j.ajo.2015.05.010 [CrossRef]
- Compera D, Entchev E, Haritoglou C, et al. Correlative microscopy of lamellar hole-associated epiretinal proliferation. J Ophthalmol. 2015;2015:450212. doi:10.1155/2015/450212 [CrossRef]
- Witkin AJ, Ko TH, Fujimoto JG, et al. Redefining lamellar holes and the vitreomacular interface: an ultrahigh-resolution optical coherence tomography study. Ophthalmology. 2006;113(3):388–397. doi:10.1016/j.ophtha.2005.10.047 [CrossRef]
- Parolini B, Schumann RG, Cereda MG, Haritoglou C, Pertile G. Lamellar macular hole: A clinicopathologic correlation of surgically excised epiretinal membranes. Invest Ophthalmol Vis Sci. 2011;52(12):9074–9083. doi:10.1167/iovs.11-8227 [CrossRef]
- dell'Omo R, Virgili G, Rizzo S, et al. Role of lamellar hole-associated epiretinal proliferation in lamellar macular holes. Am J Ophthalmol. 2017;175:16–29. doi:10.1016/j.ajo.2016.11.007 [CrossRef]
- Ko J, Kim GA, Lee SC, et al. Surgical outcomes of lamellar macular holes with and without lamellar hole-associated epiretinal proliferation. Acta Ophthalmol. 2017;95(3):e221–e226. doi:10.1111/aos.13245 [CrossRef]
- Lai TT, Chen SN, Yang CM. Epiretinal proliferation in lamellar macular holes and full-thickness macular holes: Clinical and surgical findings. Graefes Arch Clin Exp Ophthalmol. 2016;254(4):629–638. doi:10.1007/s00417-015-3133-9 [CrossRef]
- Nava U, Cereda MG, Bottoni F, et al. Long-term follow-up of fellow eye in patients with lamellar macular hole. Graefes Arch Clin Exp Ophthalmol. 2017;255(8):1485–1492. doi:10.1007/s00417-017-3652-7 [CrossRef]