Traction on the macula from the progression of a PVD can result in anatomic changes in the retina. These changes can include distortion in the contour of the foveal surface, intraretinal pseudocyst formation, elevation of the fovea from the RPE, or some combination that may result in reduced vision (Duker et al. 2013; Johnson 2010).

The IVTS defined vitreomacular traction (VMT) as the concurrence of detectable retinal anatomic changes on OCT with perifoveolar PVD (Duker et al. 2013) (Fig. 19.2). The criteria to define VMT include:

VMT can also be subclassified as FOCAL (≤1500 μm) or BROAD (>1500 μm) and ISOLATED or CONCURRENT similar to VMA (Koizumi et al. 2008). The surface area of VMT affects the anatomical distortion of the retina and the patient’s symptomatology (Fig. 19.2) (Spaide et al. 2002; Yamada and Kishi 2005). For example, focal VMT tends to distort the foveal surface, elevate the foveal floor, and form pseudocysts within the central macula that can diminish visual acuity and cause visual distortion (Haouchine et al. 2001). In contrast, broad VMT can cause a wider area of macular thickening, macular schisis, cystoid macular edema, and vascular leakage on fluorescein angiography. The SD-OCT appearance of VMT informs the optimal choice of management and the patient’s prognosis with treatment. OCT can also be used to sequentially follow VMT over time to detect resolution of the traction or, in some cases, progression to FTMH (Fig. 19.3). “Symptomatic VMA” is a clinical diagnosis with OCT showing VMT with or without concurrent macular hole and clinical complaints attributable to the retinal architectural changes (Stalmans et al. 2013). VMA and VMT may alter the pathophysiology of other retinal diseases including neovascular or wet age-related macular degeneration (wAMD) and diabetic macular edema (DME). In one meta-analysis, VMA was found to be 2.15 times more likely in the eyes with wAMD than controls and 2.54 times more likely than the eyes with dry AMD (Jackson et al. 2013). Some studies suggest that the presence of VMA impacts wAMD patients’ functional outcomes with treatment, as well as the frequency of retreatment (Waldstein et al. 2014). In patients with PVD, a lower treatment frequency of intravitreal ranibizumab may be feasible, whereas patients with VMA or recent release of vitreomacular contact may benefit from more intensive retreatment (Mayr-Sponer et al. 2013). Other studies have found an association between DME and both VMA and VMT (Jackson et al. 2013; Ophir et al. 2010). In one study, macular edema resolved spontaneously more frequently in the eyes with vitreomacular separation than in eyes with persistent vitreomacular adhesion (Hikichi et al. 1997). Asymptomatic VMA may increase the risk of diffuse DME by greater than three times as compared with diabetic patients with a complete PVD or complete vitreoretinal attachment (Lopes de Faria et al. 1999). Additional studies with SD-OCT are needed to determine the optimal treatment strategies for patients with concurrent VMA or VMT with other retinal diseases.

OCT has also led to a better understanding of the connection between PVD and the formation of epiretinal membranes. Anatomic autopsy studies have demonstrated that residual vitreous remains on the surface of the retina in nearly half of all eyes with PVD (Kishi et al. 1986). This finding called vitreoschisis can also be observed on OCT. The residual vitreous may proliferate to form an epiretinal membrane (ERM) during or after vitreous separation. This observation is supported by studies, which show a high incidence of apparent PVD in the eyes with macular pucker (Snead et al. 2008). The vitreous remnants form a scaffold for glial cells and laminocytes to attach and proliferate leading to contracture and stress on the underlying foveal architecture (Gandorfer et al. 2002).

OCT helps the physician judge whether the patient’s symptoms match the alterations in the retinal anatomy caused by ERM (Wilkins et al. 1996). This understanding allows for the most appropriate intervention given the limitations of clinical biomicroscopy. In the phase III studies of enzymatic vitreolysis with ocriplasmin, SD-OCT was found to be significantly better at detecting ERMs as compared with TD-OCT (Folgar et al. 2012). Interestingly, a clear benefit of SD-OCT over TD-OCT was not observed in the analysis of VMA or FTMHs.

There is a wide range of pathology from ERMs that can be visualized by SD-OCT (Figs. 19.4 and 19.5). A thin ERM may cause minimal alteration of the underlying retinal architecture with only a slight change in the foveal contour. However, ERMs can also result in a complete loss of the foveal contour with cystoid intraretinal fluid and marked thickening of the retina. More subtle findings on SD-OCT can include hypo-reflective cystic spaces visualized between the ERM and the internal limiting membranes (ILM).

The thickness of the ERM and degree of macular distortion and pseudocyst formation can help prognosticate the benefit of vitrectomy with membrane peel or prompt investigation of other causes for the patient’s symptoms. This is a frequently encountered situation in patients with concurrent epiretinal membranes and cataracts. The SD-OCT findings may be useful to inform whether the visual symptoms are primarily due to the cataract, the epiretinal membrane, or both and inform the management accordingly. The appearance of the bands corresponding to the photoreceptors in the outer retina (variably defined as the inner segment/outer segment (IS/OS) junction or the ellipsoid layer) is particularly useful to prognosticate visual acuity in patients undergoing vitrectomy with membrane peel or in the eyes with spontaneous release of epiretinal membranes (Inoue et al. 2011; Suh et al. 2009; Yang et al. 2014). Other studies have correlated edema of the inner nuclear layer (INL) on SD-OCT in the eyes with epiretinal membranes and metamorphopsia (Watanabe et al. 2009). These findings may be useful to the retina surgeon in determining who may benefit from removal of epiretinal membranes and in the preoperative counseling of patients.

Macular pseudohole is a clinical diagnosis made at the slit lamp with a contact or a noncontact lens. It is a clinical descriptive term for a finding that appears to be a full-thickness retinal defect, but subsequent OCT proves that the outer retina is intact. Clinically, what is most often seen is a discrete, reddish, round, or oval lesion in the fovea simulating the appearance of a full-thickness macular hole (Haouchine et al. 2004). OCT with multiple foveal line scans is critical to differentiate between a true full-thickness macular hole and a pseudohole. On OCT, the pseudohole has no loss of foveal tissue and central foveal thickness is normal or slightly thin (Haouchine et al. 2004). The IVTS further defined a macular pseudohole based on four OCT characteristics (Duker et al. 2013):

Based on the observations from biomicroscopic examination and OCT, an ERM is thought to be causative with contraction pulling the underlying retinal tissue toward the center of the fovea. This process results in invagination of the perifoveal retina into a shape that resembles a hole with no loss of tissue. Most patients with pseudohole are minimally symptomatic and can be observed. If the associated ERM causes a decrease in vision, the ERM can be removed by surgical membrane peel.

OCT is now considered the gold standard for the preoperative and postoperative management of the diseases of the VMI (Goldberg et al. 2014). OCT is necessary for an accurate diagnosis and guides preoperative decision-making and surgical planning. In one study, six vitreoretinal specialists examined patients with slit lamp biomicroscopy to diagnose ERM or VMT and determine whether surgery was indicated (Do et al. 2006). The clinical findings were then compared to the results of time domain (TD) OCT. The TD-OCT was more sensitive than the clinical exam, particularly for VMT. The higher resolution of SD-OCT further enhances the clinician’s ability to make an accurate diagnosis and manage the patient accordingly. SD-OCT is helpful to guide the management of VMT. The size of the vitreous adhesion can inform the patient’s prognosis with vitrectomy to relieve macular traction. In one study, the eyes with focal VMT had a significantly greater improvement in vision with vitrectomy compared with the eyes with broad VMT (Sonmez et al. 2008). The anatomical distortion of the retina can also inform the patient’s prognosis. In another study of patients undergoing vitrectomy for VMT, a lamellar separation between the inner and outer fovea was not associated with a significant improvement in vision, whereas cystoid macular edema and perifoveal traction were associated with improved vision (Witkin et al. 2010).

SD-OCT is also helpful to determine the thickness and extent of an ERM to inform surgical planning and predict the difficulty of ERM removal (Kim et al. 2012). Surgical difficulty of ERM removal is strongly associated with more extensive adherence as observed on SD-OCT (Kim et al. 2012). SD-OCT findings may also inform the surgeon’s decision to peel the ILM (Chang et al. 2013; Seidel et al. 2013). In one study, the greater extent of ERM elevation above the retina and thickness was predictive of ILM persistence after ERM peel (Seidel et al. 2013). Only peeling the ERM (and not ILM) results in a greater proportion of the eyes with residual ERM, yet one study revealed no differences in postoperative visual acuity between single and double peeling (Chang et al. 2013). SD-OCT can also guide the surgeon as to where to initiate the membrane peel based on the elevation above the retina, thickness, and the presence of an identifiable edge (Hirano et al. 2010; Kim et al. 2012). These insights may make the surgery easier and potentially reduce operating times. Retinal surface EN FACE OCT holds promise to further guide surgical planning of ERM by determining morphological changes on the retinal surface and areas devoid of ILM secondary to ERM contraction (Rispoli et al. 2012).

SD-OCT is also helpful to inform the patient’s prognosis with ERM peeling. Multiple studies of the eyes with ERM have demonstrated that the preoperative integrity of the IS/OS junction band on OCT helps prognosticate postoperative visual acuity (Falkner-Radler et al. 2010; Inoue et al. 2011; Suh et al. 2009). In the eyes with a disrupted IS/OS prior to surgery, none had a normal appearing IS/OS junction on SD-OCT after indocyanine green (ICG)-assisted internal limited membrane (ILM) peeling when followed up to 1 year postoperatively (Inoue et al. 2011). There was no correlation between preoperative central foveal thickness, presence of a macular pseudohole, and presence of retinal cysts with the postoperative visual acuity (Inoue et al. 2011).

Postoperative SD-OCT can also provide insights on the outcome of surgery. In particular, SD-OCT may be helpful in evaluating poor visual outcomes that may be attributable to subretinal or intraretinal fluid, disruptions to the RPE or photoreceptors, or recurrence of the ERM. These findings may guide treatment of postoperative cystoid macular edema or help determine if repeat surgery is indicated.

The approval of ocriplasmin (Jetrea, ThromboGenics, Iselin, New Jersey) for “symptomatic vitreomacular adhesion” in October 2012 opened a new era in the management of vitreoretinal interface disease (FDA 2012). Ocriplasmin is a recombinant protease with activity against components of the vitreoretinal interface including fibronectin and laminin. In phase III, MIVI-TRUST studies, 652 eyes were randomized with 464 receiving 125 μg of ocriplasmin by intravitreal injection and 188 receiving a placebo injection of saline (Stalmans et al. 2012). VMA resolved in 26.5 % of ocriplasmin-injected eyes compared to 10.1 % of placebo-injected eyes at 28 days after injection. This primary endpoint was determined by TD-OCT. Therefore, ocriplasmin is the first drug to be approved by the FDA based on an OCT analysis rather than a visual acuity or other functional outcome. Additional findings from the original trials included higher rate of closure of macular holes (40.6 % of ocriplasmin-injected eyes versus 10.6 % of placebo-injected eyes) and a higher percentage of patients with three lines of visual acuity gains (12.3 % of eyes injected with ocriplasmin versus 6.4 % of eyes injected with placebo). SD-OCT is critical to determine the ideal candidates for pharmacologic vitreolysis. In phase III trials, a higher rate of VMA resolution was found in patients without epiretinal membranes (37.4 % in the ocriplasmin group as compared with 14.3 % in the placebo group). Of patients with epiretinal membrane, VMA resolution occurred in just 8.7 % with ocriplasmin compared with 1.5 % with placebo. These data suggest that the proteolytic action of ocriplasmin is not as effective in the presence of ERMs. The careful examination of the OCT to determine if an ERM is present can inform the clinician’s choice of optimal therapy.