11 Cystoid Macular Edema
Cystoid macular edema (CME) can be broadly defined as the accumulation of excess fluid in the extracellular space of the neurosensory retina presenting clinically with cyst-like (cystic) spaces and abnormal thickening of the macula. It was first described by Irvine in 1953, 1 but Gass and Norton 2 provided the first detailed microscopic and angiographic description of this condition in 1966, and the condition later became known as Irvine–Gass syndrome. Although intracellular edema of Müller’s cells has been observed, the classic pathology of CME consists of large cystic spaces in the outer plexiform layer of Henle. 3 With modern imaging, fluid-filled spaces can often be seen in multiple layers of the retina, including the subretinal space. Although the definitive cause of CME cannot be precisely determined in most cases, several basic pathophysiologic processes contribute to the development of CME, depending on the underlying etiology and clinical context. As a final common complication of many retinal conditions, CME can be considered a leading cause of central vision loss and is, therefore, of enormous socioeconomic importance.
Therapeutic approaches to CME have evolved over the past two decades as research has led to greater understanding of underlying etiology and pathophysiologic mechanisms. The recent availability of drugs to inhibit vascular endothelial growth factor (VEGF) has been a remarkable advance in the treatment of CME due to underlying retinal vascular disease. Recent advances in drug delivery, particularly enhanced bioavailability of topically applied and intravitreally delivered anti-inflammatory drugs, have also improved treatment options.
11.2 Classification and Diagnosis
CME can be classified as clinical (retinal thickening observed on biomicroscopic examination in combination with vision impairment), angiographic (leakage detected on fluorescein angiography [FA]), and, more recently, in terms of the presence of intraretinal (with or without subretinal) fluid on optical coherence tomography (OCT). Although moderate and severe cases of CME can be observed reliably by direct biomicroscopy, mild cases may be missed. FA has been the traditional method to detect CME and can reveal a petaloid-like pattern of hyperfluorescence in the macula during the late phase due to dye leakage from perifoveal capillaries and dye accumulation within cystic spaces in the outer plexiform layer (Fig. 11-1). OCT is a relatively newer technology and provides high-resolution cross-sectional imaging that directly measures macular thickness. It has several advantages over angiography, including faster speed, noninvasiveness, and reproducibility. In addition, its measurements are quantifiable and, therefore, OCT lends itself well to clinical trials. 4 Consequently, OCT has been used increasingly to grade and assess CME in clinical practice. Despite these inherent advantages, mild diffuse cases of CME with minimal thickening may be missed on OCT but observed more readily on angiography, which shows direct leakage from retina blood vessels (Fig. 11-2).
CME is not a specific disease but rather a clinical finding of intraretinal fluid in the macula (macular edema [ME]) typically (but not always) with cystic changes that is a common complication of many diseases. Because of its historical description by Gass and Norton 2 after cataract surgery (Irvine–Gass syndrome) and its presumed relationship with inflammation, 5 CME is often used in the literature to describe ME in the setting of cataract surgery, uveitis, and other conditions (e.g., retinitis pigmentosa [RP]) where inflammation is thought to be primarily involved. In contrast, ME is more commonly used in the literature to describe similar clinical findings in the setting of retinal vein occlusions (RVOs), radiation retinopathy, and other retinovascular diseases. Furthermore, when observed in the setting of diabetes, the descriptive term diabetic macular edema (DME) is used routinely. It is now generally accepted that chronic inflammation plays a more prominent role in the pathogenesis of ME in retinal vascular diseases than recognized previously, obscuring the boundaries of this traditional classification. 6 , 7 Therefore, throughout this chapter, CME and ME are intended to be viewed as overlapping rather than distinctive terms, 8 but where appropriate the more cited term will be used for historical consistency.
CME is not a specific disease but a clinical finding which complicates many diseases. As such, its application can overlap with ME and DME. Consequently, distinguishing CME from ME or DME is not always clinically possible or even important as treatment approaches are often similar.
The volume and composition of the extracellular compartment of the neurosensory retina is tightly regulated under normal circumstances by the inner (retinal capillary endothelial cell tight junctions) and outer (retinal pigment epithelium [RPE] cell tight junctions) blood–retinal barriers and by the active pumping function of the RPE cells. 3 , 9 Consequently, cellular (endothelial or RPE) impairment/death, loss of tight junction integrity, and increased intraluminal pressure can disrupt and/or overwhelm these barriers/RPE pump capacity and result in accumulation of intraretinal fluid (see text box). 10 , 11 While vitreomacular traction also results in cystic spaces on OCT (Fig. 11-3) and in some circumstances traction-induced macular capillary leakage on FA, its pathophysiology is distinct and primarily due to mechanical traction and separation of intraretinal layers.
Main Pathophysiologic Mechanisms in Cystoid Macular Edema
Increased vascular permeability
Loss of pericytes and endothelial cells
Vascular incompetence (e.g., Coats’ disease)
Vasopermeability factors (e.g., VEGF)
Increased intraluminal pressure
Increased blood flow
Increased blood volume
Vascular occlusion (e.g., vein occlusion)
Dysfunction of RPE barrier/pump
Despite its multiple mechanisms, ME occurs most commonly due to pathologic hyperpermeability of retinal capillaries, which can occur as a result of endothelial cell damage/impairment (e.g., diabetes, Coats’ disease) or increased vascular permeability from inflammation (e.g., postsurgical, uveitis; Table 11-1). 3 Adherence of leukocytes to vessel walls (leukostasis), an early finding in diabetic retinopathy (DR) and mediated by intercellular adhesion molecule-1 (ICAM-1), 12 results in progressive endothelial damage from nitric oxide and inflammatory mediators. These mediators include prostaglandins, leukotrienes, protein kinase C (PKC), VEGF, nitric oxide, and various cytokines which result in increased retinal vascular permeability by directly or indirectly causing vasodilation and disruption of endothelial tight junctions. 13 , 14 , 15 Subsequent increased vascular permeability results in extravasation of proteins, fluid, and other macromolecules into the extracellular space and retinal interstitium. This causes a shift in the balance of hydrostatic and oncotic pressure, favoring the accumulation of fluid within the extracellular space and the development of ME. The majority of cases of ME demonstrate fluid in the outer layers of the retina, but severe occurrences can result in secondary fluid migration into the subretinal space.
Extravasation and accumulation of fluid into the retina can also be caused by disruption of the RPE barrier and/or pump function which may occur in conditions such as central serous retinopathy, retinitis pigmentosa (RP), and drug-induced toxicity and which may result in serous macular detachment and secondary ME. Retinal permeability can be increased by conditions that increase blood volume, flow, and intraluminal pressure (e.g., central retinal vein occlusion [CRVO]) or by factors that promote vasodilation.
Independent of the mechanism(s) of CME, visual acuity depends on a number of factors, including duration and severity of edema, underlying health of the RPE and photoreceptors, and macular perfusion. With the notable exception of postsurgical ME occurring in healthy retinas preoperatively, 8 many studies report poor correlation between macular thickness and visual acuity in eyes with recurrent or longstanding ME. 16
11.4 Major Pathologic Conditions
11.4.1 Intraocular Surgery
More than 2 million cataract surgeries are performed annually in the United States, and development of CME remains the most common cause of postoperative vision loss. Incidence rates of CME vary substantially throughout the literature depending on which definition is used and the type of patients included in the study. It is clear that CME occurs at a higher frequency in patients with uveitis or diabetes. 17 , 18 Recent studies have reported incidences following modern, small-incision, uncomplicated cataract surgery in healthy individuals (without diabetes or uveitis) as high as 9 to 19% using FA, but visually important CME is reported at much lower rates (in the range of 1–4%). 4 Although CME can be treated, its development increases the cost of cataract surgery by approximately 50% (additional cost in 2014: $1,092) and chronic disease can result in permanent visual impairment. 19
Although the exact pathogenesis of CME after cataract surgery remains unknown, disruption of the blood–retinal barrier due to inflammation may play a prominent role. It has been hypothesized that release of prostaglandins and other inflammatory mediators presumably from anterior uveal tissue increases permeability of perifoveal capillaries, resulting in accumulation of fluid and cystic changes in the retinal layers. 5 In agreement with this, a higher incidence of CME is seen in complicated cataract surgeries with vitreous loss, adhesions to the cataract wound, and retained lens material where inflammation would likely be more pronounced and prolonged. 20 Recent studies have confirmed higher rates of ME in diabetic eyes after cataract surgery, particularly in those eyes with a previous history of DME and, therefore, likely to have existing retinal vasculature hyperpermeability and compromise. 21 While it is virtually impossible to distinguish postsurgical CME (Irvine–Gass) versus exacerbation of DME in diabetic eyes after cataract surgery, such a distinction is mostly academic, as treatment goals and strategies for each entity are similar. Instead, emphasis should be focused on preventive measures in eyes with preexisting DME undergoing cataract surgery.
History of previous DME is significantly associated with development of ME after cataract surgery. 21 Such at-risk eyes should receive appropriate preventive measures which may consist of preoperative topical corticosteroid and/or nonsteroidal anti-inflammatory drops and periocular or intraocular corticosteroids.
Similarly, measures should be taken to ensure that inflammation is controlled for at least 3 months in eyes with uveitis before elective cataract surgery. A recent prospective trial demonstrated that administration of perioperative systemic prednisone starting 2 days before surgery with rapid taper after surgery reduced the risk of CME at 4 weeks by sevenfold. 18
Perioperative systemic prednisone significantly reduced the incidence of CME at 4 weeks after cataract surgery by sevenfold in eyes with uveitis. 18
Although much attention has focused on CME after cataract surgery, CME has also been reported after other types of intraocular surgery 22 , 23 and laser therapy. 24 , 25 , 26 New or exacerbation of CME is well known to occur following panretinal photocoagulation in diabetic eyes. 24 CME has also been reported to occur after neodymium-doped yttrium aluminum garnet (Nd:YAG) posterior capsulotomy 25 and selective laser trabeculoplasty. 26
Although there is no Food and Drug Administration (FDA)-approved therapy for CME, both prophylaxis and treatment are directed toward blocking inflammatory mediators (mainly prostaglandins) produced in the anterior uvea through the use of topical corticosteroids and nonsteroidal anti-inflammatory drugs (NSAIDs). An extensive meta-analysis of the world literature in 1998 concluded that NSAIDs are beneficial for the treatment and prevention of CME following cataract surgery. 27 A subsequent major review of the literature on this topic in 2010 reported similar findings but emphasized the paucity of well-designed studies and the lack of evidence of long-term benefit. 28 In particular, as many cases of CME are mild and resolve spontaneously, it remains unknown whether prophylactic treatment improves long-term visual outcomes.
Despite markedly increasing intraocular pressure in a small percentage of patients, corticosteroids should continue to be a mainstay of prevention and treatment of postsurgical CME, given their long track record of use and potent anti-inflammatory properties. A step-wise approach should be taken, with initial efforts focused on prevention usually with topical corticosteroids. If CME develops and is mild, increased dosing and/or prolonged use of a topical corticosteroid can be attempted. In more severe cases, local or intravitreal injection may be necessary. 29 Carbonic anhydrase inhibitors have been reported to be useful in some cases. 30 Both bevacizumab and ranibizumab have been used to treat refractory CME after surgery in uncontrolled case series with inconsistent results. 31 , 32
Despite widespread use, the benefit of topical NSAIDs in combination with topical corticosteroids in the prevention and treatment of postsurgical CME remains unknown.
Uveitis is responsible for approximately 10% of all cases of legal blindness in the United States. 33 CME is the primary cause of vision loss associated with uveitis and complicates a wide variety of uveitic syndromes, particularly posterior uveitis and panuveitis. 34 While the pathogenesis of CME is not entirely understood, resolution of CME in response to anti-inflammatory treatment strongly implicates inflammatory mediators such as prostaglandins, leukotrienes, and cytokines, with resulting compromise of the inner blood–retinal barrier. Over time, RPE barrier and/or pump function may permanently decline and contribute to persistent fluid accumulation despite adequate control of inflammation. This later occurrence is supported by the observed therapeutic benefit for uveitic CME of carbonic anhydrase inhibitors, which function primarily by facilitating fluid transport across the RPE. 35 , 36 Over the past two decades, greater emphasis has been focused on sustained suppression of chronic inflammation to prevent CME and permanent vision loss. 37 Therapeutic strategies employed rely on topical corticosteroids; adjuvant, systemic, low-dose prednisone (= 10 mg/d); and steroid-sparing immunosuppressive drugs. 38 Typically, control of ocular inflammation requires much higher doses of steroid-sparing drugs than needed to treat systemic autoimmune disease. Administration of systemic treatment with prednisone and/or a steroid-sparing agent is particularly advantageous in patients with bilateral disease and, unlike local therapy, downgrades activation and proliferation of leukocytes in peripheral lymph nodes.
Similar to postsurgical CME, treatment of more refractory or chronic cases requires a step-wise approach. There are two sustained delivery devices that are FDA approved for the treatment of uveitis. 39 , 40 A 0.7-mg dexamethasone implant (Ozurdex; Allergan, Inc.) can be injected in the office and provides sustained release of the drug for up to 6 months. Its efficacy in reducing inflammation was proven in a large randomized trial with acceptable rates of cataract formation and elevated intraocular pressure. 39 , 41 The dexamethasone implant may be particularly advantageous in vitrectomized eyes where the half-life of intravitreally injected triamcinolone acetonide is markedly reduced. A surgically implantable device containing 0.59-mg fluocinolone acetonide provides sustained release of the drug for over 2 years and was comparable to chronic systemic immunosuppression in a large randomized clinical trial. 42 Although the implant was associated with a much higher rate of ocular side effects, mainly cataract formation and intraocular pressure elevation, it controlled inflammation slightly better than systemic immunosuppression and was associated with higher quality-of-life scores. It therefore remains a viable option to control recalcitrant uveitis and CME particularly in patients intolerant of systemic immunosuppression. Treatment of uveitic CME with anti-VEGF therapy has also been reported to be beneficial in small and uncontrolled case series. 14