Purpose
To evaluate the natural course of idiopathic vitreomacular traction (VMT) with spectral-domain optical coherence tomography (SDOCT) from the vitreomacular adhesion (VMA) stage to the spontaneous resolution of VMT.
Design
Prospective observational case series.
Methods
We studied the natural course of idiopathic VMT in 46 eyes (46 patients), divided into those that proceeded to spontaneous VMT resolution (12 cases) and those that remained at the VMT stage (34 cases). All patients were examined with SDOCT at regular 3-month intervals. We recorded the vitreomacular angle of VMA nasally and temporally, the horizontal diameter of VMA, macular thickness, visual acuity, photoreceptor layer, and external limiting membrane.
Results
In the 12 eyes that proceeded to spontaneous resolution, the vitreous adhesion angle had a mean increase of 38 degrees at VMT, compared to the angle at the VMA stage. In the 34 eyes that remained at the VMT stage, the mean angle of traction increased by only 1 degree throughout follow-up. In all 46 patients, the angle at the VMT stage was significantly associated with traction resolution (nasally P = .001, temporally P < .001). The likelihood of resolution was more than 99% lower for patients with a VMT diameter >400 μm compared with that of eyes with a VMT diameter <400 μm. Patients with broad-type VMT remained at the same stage, whereas patients with V-type VMT had 80% probability of resolution.
Conclusions
Spontaneous VMT resolution is negatively associated with the horizontal adhesion diameter. The strength of the traction exerted by the vitreous on the fovea seems to be positively related to the size of the vitreomacular angle.
Posterior vitreous detachment (PVD) is a frequent consequence of aging. With age, the vitreous degenerates, leading to PVD, probably attributable to biochemical changes at the vitreous interface. The process of PVD can become pathologic in some cases, when vitreous liquefaction occurs without concomitant vitreomacular interface weakening. The condition in which the remaining attachment is to the macula is known as “vitreomacular adhesion” (VMA). VMA has emerged as a distinct entity, evident in its recent assignment to an international classification of diseases. Using optical coherence tomography (OCT), VMA is characterized as the adhesion of the posterior hyaloid to the center of the fovea, with detached vitreous around the macula. At the VMA stage, the foveal contour is normal, with no distortion. However, in a subset of patients, the tractional effect in the early stages of VMA produces a variety of pathologic macular features, determined in part by the size and strength of the residual vitreoretinal adhesion. In these cases, VMA progresses to vitreomacular traction (VMT), which is characterized by the steep slope of the inner macular surface or by sharp angulation and localized deformation of the retina at the VMA site. VMT is now believed to be associated with a broad spectrum of maculopathies, including cystoid macular edema, epiretinal membrane (ERM), lamellar macular hole (LMH), and the formation of full-thickness macular hole (FTMH) with visual disturbance. In both clinical entities (VMA and VMT), the vitreous is detached around the macula but remains attached to the fovea, the optic disc, and the retinal periphery. In some cases, there is spontaneous resolution of VMT if the vitreous detaches from the fovea. In this study, we prospectively investigated the spontaneous resolution of VMT, describing all the stages of vitreomacular abnormalities from VMA through VMT to the spontaneous resolution of vitreous traction.
Methods
This was a prospective multicenter observational case series study. The study was performed at the 2nd Department of Ophthalmology of the University of Athens, the 2nd Department of Ophthalmology of Henry Dunant Hospital, and the Ophthalmological Institute of Athens. The study was conducted according to the Declaration of Helsinki and all participants gave their informed consent after the purpose of the study had been explained to them. The medical ethics committees of the participating hospitals approved the study. The Institutional Review Boards of Attikon Hospital University of Athens and Henry Dunant Hospital also approved the study. The study focused on the natural course of VMT in 12 patients (12 eyes), investigating the process of spontaneous vitreous release from VMA to VMT and finally to the spontaneous separation of the posterior hyaloid from the fovea. The 12 eyes that resolved spontaneously were compared with the 34 eyes (34 patients) in which VMT remained at that stage throughout the follow-up period. Originally, 7251 consecutive individuals were examined with OCT for various retinal pathologies. Two hundred twenty-one subjects were found to have idiopathic VMA, with no other retinal pathology or visual disturbances apart from floaters and flashing lights in 42 cases. One hundred eighty-two of these 221 VMA patients were available for further examination every 3 months from January 2010 to February 2013 in the same hospitals.
All patients were examined with Spectralis spectral-domain OCT (SDOCT) (Heidelberg Engineering, Heidelberg, Germany) imaging through dilated pupils. The horizontal B-scan images were extracted from 6 radial lines 3 mm long, scanned through the center of the fovea. For each patient, 6 radial scans 3 mm long were performed at equally spaced angular orientations centered on the foveola. We confirmed that the scan intersection point was consistent over time (Spectralis OCT with 7 μm axial resolution). Repeated scans were made at 3-month intervals by the same experienced ophthalmologist. In cases in which OCT could not identify the status of the vitreous, B-scan ultrasonography was also performed.
At the baseline examination, all eyes with VMA were characterized by the incomplete separation of the posterior vitreous, with persistent vitreous attachment at the fovea and the optic disc. It was not possible to precisely identify the onset of VMA. The foveal contour and macular thickness were normal. At the stage of VMT, there was deformation of the foveal contour, with intraretinal cysts and visual disturbance. All 46 patients included in the study had no abnormalities other than VMA and VMT. Patients with ocular trauma, with previous operations, or with other fundus diseases were excluded. We recorded the visual acuity (VA) at each follow-up visit as the logarithm of the minimal angle of resolution (logMAR). We examined the time period in which each patient remained at the VMA stage and at the VMT stage, the macular thickness, the inner segment/outer segment (IS/OS) layers of the photoreceptors, and the external limiting membrane (ELM). We also manually measured the angle of vitreous adhesion at the fovea, nasally and temporally, which was defined as the angle between the partially detached posterior hyaloid and the retinal pigment epithelium (RPE), as well as the horizontal diameter of the VMA. Our patients attended a follow-up visit every 3 months. Although we could not know the exact time at which each patient moved from one stage to the next, we calculated the time at each stage based on the cumulative observation time of 3-month intervals at each stage, which closely approximated the actual time period.
Statistical Analysis
All data were expressed as the means ± standard deviations (SD) of the means. A paired Student t test was used to compare the patient characteristics and OCT measurements between stages. Prognostic factors for resolution of the VMT stage were established with univariate logistic regression. All statistical analyses were performed with the SPSS statistical package (IBM Corp. Released 2011 IBM SPSS Statistics for Windows, version 20.0), Armonk, New York, USA) and the R statistical package (R Development Core Team 2013 R Foundation for Statistical Computing, version 3.0.1, Vienna, Austria). P values less than .05 were deemed to be statistically significant.
Results
During the first observation period between January 2010 and December 2010, 69 of 182 eyes with VMA progressed to VMT, 21 eyes developed complete vitreous separation, and 92 eyes remained at the VMA stage. In the last phase of follow-up (January 2011 to February 2013), of the 92 eyes with VMA, 14 had developed complete vitreous separation, whereas of the 69 patients with VMT, 8 had developed ERM, 5 had undergone a cataract operation, 2 had developed FTMH, and 1 had developed LMH. Seven patients underwent vitrectomy; 6 of them had ERM, reduced VA, and metamorphopsia and were willing to undergo surgery, while 1 patient had the surgery to remove a dropped nucleus. Spontaneous resolution of VMT with complete PVD occurred in 12 eyes, whereas 34 patients remained at the VMT stage. The present study includes the 12 patients with spontaneous VMT resolution and compares them with the 34 patients that remained at the VMT stage during follow-up. The study began when all the patients (46 eyes) were at the VMA stage.
The clinical characteristics of the 12 patients with spontaneous resolution are shown in Table 1 . In the group with spontaneous resolution, the mean age of the patients was 66.5 ± 6.2 years. Five patients were male (42%) and 7 were female (58%). The mean VA was 0.14 ± 0.06 logMAR at the VMA stage, 0.30 ± 0.11 logMAR at the VMT stage, and 0.19 ± 0.10 logMAR at the final outcome visit, 3 months after the complete spontaneous resolution of VMT. The difference in VA between the VMA and VMT stages was statistically significant ( P < .001). At the VMA stage, the mean angle was 12.58 ± 3.97 degrees nasally and 12.75 ± 2.99 degrees temporally, whereas at the VMT stage, the mean angles were 50.58 ± 7.66 degrees and 51.25 ± 2.99 degrees, respectively. The differences between the angles at the VMA and VMT stages were statistically significant ( P < .001). The mean horizontal diameter of the vitreous attachment at the fovea was 191.67 ± 122.17 μm, ranging from 180-420 μm. The mean macular thickness at the VMT stage was 419.17 ± 88.37 μm, and 3 months after spontaneous resolution, it was 276.08 ± 76.03 μm ( P < .001). At the final examination after spontaneous resolution and complete PVD, 7 eyes had increased macular thickness, whereas the foveal contour was restored in the remaining 5 eyes. At the VMT stage, interruption of IS/OS was noticed in 4 cases, whereas at the final examination, interruption was evident in 3 cases. The ELM was intact in all eyes in the VMT and spontaneous resolution stages. The average total follow-up of the patients was 15.8 months (±8.4), whereas the average observation period of patients at the VMA stage was 7 ± 2.95 months and at the VMT stage before spontaneous resolution was 8.75 ± 6.06 months. During the VMT stage, 1 of the 12 eyes developed ERM on the fovea, which coexisted with anteroposterior macular traction. All 12 eyes had a V-shaped configuration and were the focal VMT type, with a maximum horizontal vitreomacular attachment of 420 μm or less. OCT examinations of the fellow eyes revealed 5 eyes with V-type VMT, 2 eyes with broad-type VMT, 1 eye with VMT and epiretinal membrane, and 4 eyes with complete PVD.
| Characteristics | Vitreomacular Adhesion Stage Mean (SD) | Vitreomacular Traction Stage | Spontaneous Resolution (Final Outcome) | ||
|---|---|---|---|---|---|
| Mean (SD) | P Value | Mean (SD) | P Value | ||
| Age (y) | 66.5 (6.2) | ||||
| Male/female (%) | 42/58 | ||||
| Visual acuity (logMAR) | 0.14 (0.06) | 0.30 (0.11) | <.001 | 0.19 (0.10) | .048 |
| Angle between vitreous surface and nasal retina (degrees) | 12.58 (3.97) | 50.58 (7.66) | <.001 | ||
| Angle between vitreous surface and temporal retina (degrees) | 12.75 (2.99) | 51.25 (2.99) | <.001 | ||
| Diameter of vitreous surface adhesion (μm) | 191.67 (122.17) | ||||
| Time at stage (months) | 7 (2.95) | 8.75 (6.06) | .206 | ||
| Macular thickness (μm) | 419.17 (88.37) | 276.08 (76.03) | <.001 | ||
| IS/OS interruption, n (%) | 4 (33) | 3 (25) | |||
The clinical characteristics of the 34 eyes that remained at the VMT stage, at VMA and at the baseline and final VMT-stage examination, are shown in Table 2 . The mean age of the 34 patients was 65.3 ± 5.4 years. Fourteen patients (41%) were male and 20 (59%) female. The VA at the VMA stage was 0.13 ± 0.04 logMAR, whereas at the first and last examinations of the VMT stage, it was 0.27 ± 0.1 logMAR and 0.29 ± 0.1 logMAR, respectively ( P < .001). At the VMA stage, the mean angle was 11.75 ± 4.2 degrees nasally and 12.02 ± 3.7 degrees temporally, whereas at the baseline VMT stage, the mean angles were 29 ± 10.5 degrees and 31.4 ± 10.3 degrees, respectively ( P < .001), and at the last examination, the mean angles were 30 ± 10.5 degrees and 32 ± 10 degrees, respectively. All but 3 eyes had the broad-type configuration, with a mean horizontal vitreomacular attachment of 643 ± 242 μm, ranging from 400-1000 μm. The mean time at the VMA stage was 6.88 ± 3.18 months and at the VMT stage was 31.12 ± 3.18 months ( P < .001). The mean macular thickness at baseline was 387 ± 93.8 μm and 410 ± 100.6 μm at the final examination. In 6 of the 34 eyes, disruption of the IS/OS layer was observed at the baseline and last examinations, whereas the ELM was intact in all eyes. At the last examination, ERM was evident in 5 of the 34 eyes on OCT, which had not been detected at the first examination. OCT examinations of the fellow eyes of these 34 patients revealed 21 eyes with broad-type VMT, 3 eyes with a V-type configuration, 5 eyes with complete PVD, and 6 eyes with VMT and ERM.
| Characteristics | Vitreomacular Adhesion Stage, Mean (SD) | Vitreomacular Traction Stage | Difference, Mean ( P Value) b | ||
|---|---|---|---|---|---|
| Baseline, Mean (SD) | P Value a | Last Examination, Mean (SD) | |||
| Age (y) | 65.3 (5.4) | ||||
| Male/female (%) | 41/59 | ||||
| Visual acuity (logMAR) | 0.13 (0.04) | 0.27 (0.1) | <.001 | 0.29 (0.1) | 0.02 (.008) |
| Angle between vitreous surface and nasal retina (degrees) | 11.75 (4.2) | 29 (10.5) | <.001 | 30 (10.5) | 1.06 (.007) |
| Angle between vitreous surface and temporal retina (degrees) | 12.02 (3.7) | 31.4 (10.3) | <.001 | 32 (10) | 0.62 (.14) |
| Diameter of vitreous surface adhesion (μm) | 631.2 (137) | 643 (242) | 11.77 (.07) | ||
| Time at stage (months) | 6.88 (3.18) | 31.12 (3.18) | 24.24 (<.001) | ||
| Macular thickness (μm) | 387 (93.8) | 410 (100.6) | 22.94 (.01) | ||
| IS/OS interruption, n (%) | 6 (18) | 6 (18) | |||
a P calculated between vitreomacular adhesion stage and baseline examination.
In a comparison of those 34 eyes with the 12 eyes that underwent spontaneous vitreous separation, we found no difference in the angle of vitreous adhesion, measured nasally or temporally, in the VMA stage. However, there was a difference in the angle of vitreous adhesion. The mean angle nasally was 50.58 ± 7.66 degrees in the eyes with VMT resolution and 30.12 ± 10.56 degrees in the eyes that remained at the VMT stage ( P < .001); mean angle temporally was 51.25 ± 2.99 degrees in the eyes with VMT resolution and 32.00 ± 10.09 degrees in the eyes that remained at the VMT stage ( P = .001). We also found a difference in the diameter of the vitreomacular adhesion (mean diameter 191.67 ± 122.17 μm in the eyes with spontaneous resolution and 659.12 ± 171.05 μm in the group that remained at the VMT stage, P < .001) and the type of adhesion. All eyes with spontaneous vitreous separation were of the V type, whereas in the eyes that remained at the VMT stage, 3 of 34 eyes (9%) were of the V-type configuration and the rest were of the broad type.
Results
During the first observation period between January 2010 and December 2010, 69 of 182 eyes with VMA progressed to VMT, 21 eyes developed complete vitreous separation, and 92 eyes remained at the VMA stage. In the last phase of follow-up (January 2011 to February 2013), of the 92 eyes with VMA, 14 had developed complete vitreous separation, whereas of the 69 patients with VMT, 8 had developed ERM, 5 had undergone a cataract operation, 2 had developed FTMH, and 1 had developed LMH. Seven patients underwent vitrectomy; 6 of them had ERM, reduced VA, and metamorphopsia and were willing to undergo surgery, while 1 patient had the surgery to remove a dropped nucleus. Spontaneous resolution of VMT with complete PVD occurred in 12 eyes, whereas 34 patients remained at the VMT stage. The present study includes the 12 patients with spontaneous VMT resolution and compares them with the 34 patients that remained at the VMT stage during follow-up. The study began when all the patients (46 eyes) were at the VMA stage.
The clinical characteristics of the 12 patients with spontaneous resolution are shown in Table 1 . In the group with spontaneous resolution, the mean age of the patients was 66.5 ± 6.2 years. Five patients were male (42%) and 7 were female (58%). The mean VA was 0.14 ± 0.06 logMAR at the VMA stage, 0.30 ± 0.11 logMAR at the VMT stage, and 0.19 ± 0.10 logMAR at the final outcome visit, 3 months after the complete spontaneous resolution of VMT. The difference in VA between the VMA and VMT stages was statistically significant ( P < .001). At the VMA stage, the mean angle was 12.58 ± 3.97 degrees nasally and 12.75 ± 2.99 degrees temporally, whereas at the VMT stage, the mean angles were 50.58 ± 7.66 degrees and 51.25 ± 2.99 degrees, respectively. The differences between the angles at the VMA and VMT stages were statistically significant ( P < .001). The mean horizontal diameter of the vitreous attachment at the fovea was 191.67 ± 122.17 μm, ranging from 180-420 μm. The mean macular thickness at the VMT stage was 419.17 ± 88.37 μm, and 3 months after spontaneous resolution, it was 276.08 ± 76.03 μm ( P < .001). At the final examination after spontaneous resolution and complete PVD, 7 eyes had increased macular thickness, whereas the foveal contour was restored in the remaining 5 eyes. At the VMT stage, interruption of IS/OS was noticed in 4 cases, whereas at the final examination, interruption was evident in 3 cases. The ELM was intact in all eyes in the VMT and spontaneous resolution stages. The average total follow-up of the patients was 15.8 months (±8.4), whereas the average observation period of patients at the VMA stage was 7 ± 2.95 months and at the VMT stage before spontaneous resolution was 8.75 ± 6.06 months. During the VMT stage, 1 of the 12 eyes developed ERM on the fovea, which coexisted with anteroposterior macular traction. All 12 eyes had a V-shaped configuration and were the focal VMT type, with a maximum horizontal vitreomacular attachment of 420 μm or less. OCT examinations of the fellow eyes revealed 5 eyes with V-type VMT, 2 eyes with broad-type VMT, 1 eye with VMT and epiretinal membrane, and 4 eyes with complete PVD.
