Summarize the surgical and nonsurgical management strategies for diabetic eye disease and identify the objectives and indications for surgery.

Describe the role of vitrectomy in the management of disorders such as nonclearing vitreous hemorrhage, traction and tractional rhegmatogenous retinal detachment, progressive fibrovascular proliferation, and iris neovascularization.

Evaluate data from surgical trials of vitrectomy for diabetic macular edema.

Treat neovascular age-related macular degeneration with anti-vascular endothelial growth factor agents and photodynamic therapy.

Choose the appropriate treatment for the perfused and ischemic forms of branch vein and central vein occlusion.

Surgery for Diabetic Eye Disease: The State of the Art
Julia A. Haller, MD, Professor and Chair, Department of Ophthalmology, Jefferson Medical College, Thomas Jefferson University, and Ophthalmologist-in-Chief, Wills Eye Institute, Philadelphia, PA

Prevalence: among adults in United States, 49% increase in diabetes mellitus (DM), from 4.9% in 1990 to 7.3% in 2000; 40.3% of patients >40 yr of age with DM have diabetic retinopathy (DR), which threatens vision in 8.2%

Nonsurgical treatment: systemic control of hyperglycemia, blood pressure, and lipids; laser photocoagulation gold standard for treatment of diabetic macular edema (DME) and proliferative DR; new options include intravitreal steroids and anti-vascular endothelial growth factor (anti-VEGF) agents

Diabetic vitrectomy: objective—correct anatomic abnormalities and optimize physiologic function by removing axial opacities, relieving anterior-posterior and tangential traction, and reattaching retina; indications—nonclearing vitreous hemorrhage; tractional retinal detachment involving macula; tractional rhegmatogenous retinal detachment; eyes with rubeosis iridis and vitreous hemorrhage; some cases of progressive fibrovascular proliferation

Nonclearing vitreous hemorrhage: early intervention important (especially in patients with juvenile-onset DM) because of rapid progression; surgery involves removal of axial opacity and circumscription of posterior vitreous detachment with traction at optic nerve head; surgery highly successful

Traction retinal detachment: involves circumscribing retinal adhesion and removing tangential traction by segmenting and delaminating membranes; for bimanual technique, elevate fibrovascular plaque with lighted pick and cut off pegs; removal of anatomic problem may not fully rehabilitate ischemic retina

Combined traction/rhegmatogenous retinal detachment: more challenging because of mobility of retina; diagnosed by presence of hydration lines, concavity of retina, detachment extending into periphery, and break adjacent to area of maximum traction

Progressive fibrovascular proliferation: surgery indicated when eye resists laser treatment and continues to evolve aggressive fronds

Iris neovascularization (NV): media opacity prevents laser penetration; vitrectomy and laser treatment performed simultaneously

Expanded surgical objectives: instrumentation—improvements include small-gauge surgery, multifunctional probes, and wide-field viewing systems; procedures—increase retinal oxygenation; modulate factors affecting retinal and vascular function (eg, VEGF); improve function of optic nerve and macula

Emerging nonstandard indications: vitreopapillary traction; DME—useful for patients with persistent form; vitrectomy enables delivery of more oxygen to retina by ciliary body; removal of posterior hyaloids and peeling of epiretinal membrane important if traction present; benefit of peeling internal limiting membrane unproven; intravitreal steroids used to stain membranes, define tissue planes, and decrease edema; peripheral scatter photocoagulation may help decrease production of VEGF; limited data available on benefits and risks; few natural history data on consequences of not treating vitreomacular traction

Diabetic Retinopathy Clinical Research Network

Trial characteristics: used cohort design to evaluate vitrectomy for DME; most patients did not have traction; enrollment criteria included DME and visual acuity (VA) better than 20/400; procedure chosen by surgeon; patients evaluated at 3 and 6 mo; primary traction cohort had traction, VA of 20/63 to 20/400, edema >300 µm on optical coherence tomography (OCT), and absence of cataracts; 241 eyes enrolled, 87 in primary traction cohort

Outcomes: maculae decreased from 500 to 250 µm (only 50% decreased below 250 µm); at 6 mo, 37% of patients had significant improvement (10 letters) in VA, and 9% improved by 1 or 2 lines; 25% had 2-line decrease; mean VA relatively unchanged

Complications: vitreous hemorrhages (5%), elevated intraocular pressure (IOP) requiring treatment (few cases), retinal detachment (1%), and endophthalmitis (1 patient)

Natural history of vitreomacular traction: retrospective study conducted in 53 eyes of patients without DM; at 6 mo, twice as many patients (30%) had VA decrease to 20/200 or worse; 34 (64%) had 2-line decrease in VA at time of final visit (6-60 mo); unless spontaneous posterior vitreous detachment (PVD) developed, 87% worsened at final vision

Subgroup analysis: not statistically significant, but younger patients had improvement of 9 letters, which correlated with greater improvement in OCT results; patients with shorter duration of DM, DME, and traction may have done better

Ongoing evaluations: efficacy seen in patients with internal limiting membrane peeling; benefits of adjunctive pharmacologic agents (eg, SurModics device, dexamethasone implant [Posurdex], and fluocinolone implant [Medidur]); use of vitreolytic enzymes (eg, chondroitinase) to supplant or aid vitrectomy

Update on Management of Age-related Macular Degeneration (AMD)
Henry A. Leder, MD, Assistant Chief of Service, Wilmer Ophthalmological Institute, Johns Hopkins University School of Medicine, Baltimore

Background: 90% of AMD non-neovascular form, but most of vision loss results from neovascular or wet form, characterized by choroidal neovascular membranes, subretinal fluid, leakage, and hemorrhages

Vascular endothelial growth factor: implicated in DR, DME, central retinal vein occlusion (CVO) and branch retinal vein occlusion (BVO), formation of choroidal neovascular membrane, and AMD; stimulates angiogenesis, increases vascular permeability and fenestrations, and mediates inflammatory response; also has neuroprotective function

VEGF family: includes 5 subtypes, ie, VEGF-A, -B, -C, -D, and placental growth factors (PIGF)-1 and 2; induced by hypoxia, nitric oxide, and other growth factors; therapeutic agents target VEGF-A, which consists of isoforms 121, 165, 189, and 205; pegaptanib (Macugen) targets isoform 165 specifically; nonspecific VEGF-A inhibitors more successful

Treatment of NV-AMD: Macular Photocoagulation Study (MPS) used laser successfully but unsuitable for patients with subfoveal and juxtafoveal lesions; photodynamic therapy (PDT) with verteporfin maintained but did not improve vision; anti-VEGF treatment now standard of care

Ranibizumab (Lucentis): antibody fragment binds all isoforms of VEGF-A, with shorter half-life in vitreous and serum than bevacizumab (Avastin); approved for NV-AMD

Clinical trials: Minimally classic/occult trial of the Anti-VEGF antibody Ranibizumab In the treatment of Neovascular AMD (MARINA)—patients received monthly ranibizumab or placebo; 25% to 34% of patients had vision improved by 15 letters on Early Treatment of DR study (ETDRS) chart (3 lines); 95% of patients retained vision after 1 yr, compared to 62% in placebo arm; Anti-VEGF antibody for the treatment of predominantly classic CHORoidal neovascularization in AMD (ANCHOR) trial—compared ranibizumab to verteporfin in patients with predominantly classic lesions; 35% to 40% of patients had improved vision, compared to 5.6% with PDT; 95% (vs 64%) retained vision

Frequency of injection: monthly injections expensive and increase risk for complications (eg, endophthalmitis); 3-mo induction phase used in most trials; Phase IIIb, multicentre, randomized, double-masked, sham Injection-controlled study of Efficacy and safety of Ranibizumab (PIER) study—evaluated injection every 3 mo after induction phase; Safety Assessment of Intravitreal Lucentis for AMD (SAILOR) trial—injections given according to circumstances (“as needed”); average VA decreased in both studies

Combination therapy: EVEREST, DENALI, and MONTBLANC studies—evaluated combination of ranibizumab and PDT in the treatment of choroidal NV due to AMD

DENALI: patients randomized to 1 of 3 arms, ie, monthly injections of ranibizumab, or ranibizumab combined with verteporfin at either standard fluence or reduced fluence; treatment administered as needed after 3-mo induction

Bevacizumab: Comparison of Age-Related Macular Degeneration Treatment Trials (CATT); noninferiority trial with 1200 patients; 4 arms (2 with monthly dosing of either ranibizumab or bevacizumab, and 2 with 3-mo induction phase, then as-needed dosing)

VEGF-Trap-Eye: intraocular injection of fusion protein of human VEGF receptors 1 and 2 combined with Fc portion; higher affinity for VEGF than monoclonal antibodies; blocks all isoforms of VEGF-A and PIGF; penetrates layers of retina

CLEAR-IT study: phase II trial with 5 arms, ie, 3 doses (0.5-, 2-, and 4-mg injections) and 2 dosing schedules; outcomes—all arms showed increased VA at 4 and 8 wk of treatment; by 12 wk, VA decreased in group that received injections only at 0 and 12 wk; best VA achieved in groups with injections every 4 wk; 2-mg dose better than 0.5-mg dose; rapid improvement in central thickness seen, with some regression at 12 wk; greatest improvement seen in group receiving 4 mg; next best in group with 2 mg every week; as-needed dosing phase—patients received mean of 1 injection (range 1.35 to 0.65) over 20 wk; 55% of patients receiving 2-mg injections every 4 wk received no injections during as-needed dosing phase; 32-wk results—VA ranged from 88% to 100%

VEGF-trap Investigation of Efficacy in Wet AMD (VIEW) study: phase III comparison of VEGF-Trap-Eye vs ranibizumab; 4 arms with different dosing schedules

Sirolimus (Rapamune): binds FK-binding protein (FKBP12) and targets mTOR (protein kinase that regulates proliferation, angiogenesis, and fibrosis); approved for prevention of graft rejection, restenosis of coronary grafts, and treatment of renal cell carcinoma; prevents angiogenesis by decreasing production of and response to VEGF-A; blocks hypoxic response and stops fibrosis

AMD trial: phase I study of 10 patients randomized to intravitreal or subconjunctival injection with dose escalation; VA increased in both groups, then decreased by 90 days; central foveal thickness decreased in both groups over duration of study

Emerging treatments: short interfering RNA (siRNA); tyrosine kinase inhibitors; nicotinic acetylcholine receptor antagonists; integrin antagonists

Management of Retinal Vein Occlusion
Daniel Finkelstein, MD, Professor, Department of Ophthalmology, Wilmer Ophthalmological Institute, Johns Hopkins University School of Medicine

Branch vein occlusion: laser scatter photocoagulation to ischemic zone—prevents NV by 50%; reduces vitreous hemorrhage by 50% if NV already present; useful only after NV has appeared

Perfused ME: characterized by dilated capillaries in segmental pattern and leakage into center of fovea; distinct from ischemic edema (very pronounced and cystoid, with capillary dropout and little leakage); laser management—used only for perfused ME; grid photocoagulation shown to improve VA by 1 to 2 lines in 66% of patients; 33% improve without laser treatment; 1 to 2 lines of VA seen on average

Recommendations: laser grid photocoagulation recommended for perfused ME, not for ischemic edema (edema always spontaneously improves, although VA may not); wait 6 to 12 mo to clear hemorrhage, evaluate VA, and determine ischemia before performing scatter or grid laser photocoagulation; discuss long-term prognosis with patient during waiting period; if VA 20/40 or worse from perfused edema, consider grid photocoagulation; if NV present, scatter photocoagulation recommended

Central vein occlusion: must distinguish perfused form (ie, mild, incipient, impending, nonischemic, or nonhemorrhagic) from ischemic form (ie, nonperfused, severe, or hemorrhagic); perfused form—rubeosis unlikely; VA 20/200 or better, with less hemorrhage; vascular characteristics observable by fluorescein angiography (FA); ischemic form— rubeosis possible, VA worse than 20/200 with more hemorrhage (blood and thunder or tomato catsup fundus), and FA not usually possible

Treatment: follow patients with perfused form every 2 mo (25% become ischemic); evaluate patients with ischemic form monthly for iris NV (50% develop iris NV in 3 mo, usually at papillary border, and 50% never develop iris NV); no prophylactic treatment recommended; if much retinal hemorrhage present, consider pan cryocoagulation

Perfused ME: 3 patients improved after laser grid photocoagulation (all <65 yr of age); with ME lessened, VA improved, cases stable for 2 yr, and VA returned to 20/30 to 20/40; clinical trial showed for most patients >65 yr of age, grid laser photocoagulation reduced edema but did not improve VA; laser anastomosis treatment (retina to choroid)— occasionally works well for cases of persistent ME; modified procedure (50 µm spot, 0.5 sec, and 1.1 watt) also successful; investigational technique requiring high-power argon; associated with choroidovitreal NV

Discussion with patients: meet often; discuss patients’ support systems, including spirituality