Ethics in the Practice of Retina

Put the patient’s interest first.


Ethics and morals are often viewed interchangeably but are, in fact, quite different. Although both concepts deal with distinguishing between what is right and wrong, morals are typically personal, while ethics are typically determined by a group or community. It is possible for one’s moral beliefs to conflict with existing ethical standards. This article will discuss ethics in ophthalmology as a standard determined by a group (such as the American Academy of Ophthalmology, the American Society of Retina Specialists, or the American Society of Cataract and Refractive Surgery) and often codified in rules and proceduralized to enable review and potential enforcement. Retina specialists often speak of a code of ethics that describes appropriate actions.


The field of retina abounds with ethical issues that are managed on a daily basis. Much of ethical ophthalmology hinges on focusing always on what is best for the patient. Interestingly, different clinicians may provide very dissimilar care to a patient but may actually be practicing ethically and in the best interest of that patient. Retina specialists exhibit great diversity in the approach to management of the individual patient. Lacking a single standard of care requires the retina specialist to incorporate, and apply, clinical acumen, appropriate clinical trial results, awareness of differing management approaches, and the individual retina specialist’s own abilities to deliver the best care to each unique patient.

Retina specialists are reminded each day how varied the clinical approaches are to the most common retinal diseases. Each day presents conflicts about which anti-VEGF to choose for neovascular AMD, whether intravitreal steroids play a role in diabetic macular edema, if surgery should be offered for symptomatic floaters, how long to position patients after macular hole surgery, and when to provide scleral buckling surgery for retinal detachment. Ethical issues are the focus of this article, which will explore these conflicts by focusing on the most at-risk patients: premature infants with retinopathy of prematurity (ROP) and adult patients with uveal melanoma.


One of the most controversial treatments in ophthalmology has been anti-VEGF for ROP. Despite the knowledge that ROP is a VEGF-driven disease, the use of anti-VEGF in the treatment of ROP has truly polarized the field.

Retinopathy of prematurity is one of the leading causes of vision loss and blindness in the world, with a worldwide prevalence of blindness of 50,000 infants. Because of the increasing survival of low-birth-weight infants the absolute number of infants with ROP has increased,1 and with greater at risk babies, blindness from ROP will continue to occur without advances in treatment.

Retinopathy of prematurity is a difficult disease to diagnose and treat. The disease requires the expertise of the subspecialized ophthalmologist not only to screen but to treat. Laser photocoagulation has been the standard of care, but laser photocoagulation has its pitfalls. It requires an expensive machine, it is difficult to perform without complications, and there is often progression of disease from suboptimal laser treatment. To find expertise in laser photocoagulation for ROP, both nationally and internationally, and to have expensive equipment available for these infants is quite challenging. Therefore, vision loss from ROP has remained the number one cause of childhood blindness in the world. The introduction of anti-VEGF treatment, at an international level, has been a true revolution in ROP care, markedly reducing blindness from treatment-warranted ROP.

In 2011, the results of the first randomized, multicenter, prospective trial for the use of anti-VEGF in ROP were released. The BEAT-ROP study (Bevacizumab Eliminates the Angiogenic Threat of Retinopathy of Prematurity) showed that intravitreal bevacizumab (0.625 mg in 0.025 mL) in eyes of babies with stage 3+ zone 1 or posterior zone 2 ROP resulted in lower rates of recurrence compared to conventional laser (26% vs 6%, P<.05).2

The international ophthalmology community adopted the use of anti-VEGF for ROP much more quickly than did US retina specialists. Internationally, it was noted that anti-VEGF was more accessible financially to the physicians and families of these complex premature infants and, just as importantly, it was easier to deliver than laser photocoagulation. Furthermore, the ROP disease seen in developing countries seems to respond well, in almost all cases, to a single injection of anti-VEGF. Anti-VEGF clearly saves eyes with ROP that before would have progressed to being untreatable and inoperable. The international community quickly understood that anti-VEGF was better than any other treatment when considering ease of administration, relative cost, and, most importantly, improved anatomic and functional outcomes. Further, using anti-VEGF improved outcomes even for babies in whom progression occurred or for babies who were lost to follow-up. Ultimately, the chances of having a viable macula and formed vision was greater with anti-VEGF treatment than with no, or inadequate, laser treatment.

In the United States, the concern that has driven the slow adoption of anti-VEGF usage at many academic institutions has been the potential for systemic or neurocognitive developmental delay after treatment with anti-VEGF. This theory has not been proven in scientific literature, and the international community has not reported evidence that it is true. There has been awareness among those treating ROP and desire to evaluate these concerns, but there has been no proof that anti-VEGF damages development. These concerns have prompted appropriate discussions of the ideal drug, dose, and delivery schedule for treatment-warranted ROP. There are continued arguments that ranibizumab or a lower dose of bevacizumab could be a better treatment approach. These discussions are tempered by preliminary studies that have shown earlier recurrence of disease with ranibizumab vs bevacizumab and also higher failure rates with reduced dosing of bevacizumab.

From a neurocognitive perspective, ophthalmologists have always believed that impairment of visual function, when all else is equal, truly limits the developmental potential of a child. Premature infants with ROP have overwhelming comorbidities that are not amenable to simple treatments such as a single injection of anti-VEGF for treatment-warranted ROP.

In this case, the known benefit of anti-VEGF is actively weighed against the potential concerns for negative systemic effects for these children. For many ROP specialists, the reality is that anti-VEGF is the best treatment for a VEGF-driven disease, especially when it is a one-time treatment to control abnormal development. How can a retina specialist ethically treat a child who has aggressive posterior ROP using a destructive treatment that will most likely doom them to legal blindness and has a high potential for future retinal complications? From a patient care perspective, this is unacceptable.


Uveal melanoma is the most common primary intraocular tumor of adults. Typically, this malignancy presents asymptomatically during the 6th decade. Previous studies have suggested a melanoma-specific mortality rate of 50% for patients treated with either enucleation or radiotherapy. In the National Eye Institute/National Cancer Institute’s randomized clinical trial, the 5-year melanoma-specific mortality rate for eligible medium tumor trial patients was approximately 10%. The Collaborative Ocular Melanoma Study (COMS) nonrandomized, noncontrolled study of small melanoma patients reported the 5-year mortality rate as 6%.3 Unfortunately, mortality increased significantly to 15% by 8 years. In this study, approximately one-third of patients experienced tumor growth.

Recently, tumor molecular genetics have become a significant prognostic factor in development of metastatic disease and tumor mortality. Genetic expression profiling (GEP) testing describes a 5-year metastatic rate of 72% in patients with class 2 tumors, while clinically similar tumors with class 1A GEP have a 5-year metastatic rate of 2%.4 Over the last 3 decades, a major shift away from enucleation as a primary treatment of uveal melanoma has occurred, with a focus on globe salvage using radiotherapy.5 Radiotherapy-treated eyes developed severe limitations in visual function (20/200 or worse), typically by 36 months, that were associated with radiation maculopathy, radiation optic neuropathy, or radiation-induced neovascular glaucoma. Radiation complications are now amenable to anti-VEGF therapy, and recent studies have suggested that with early, aggressive anti-VEGF treatment, mean VA at 5 years approximates 20/50 (a major improvement in visual function), while secondary enucleation rates have declined from approximately 15% after radiotherapy to virtually zero. Finally, diagnostic accuracy for COMS medium tumors between 2.5 mm and 3.0 mm was shown to be virtually 100% in enucleated eyes strongly confirming the ability of ocular oncologists to recognize smaller malignant melanomas.

The ethical dilemma resides in the ubiquitous risk and benefit assessment for treatment of uveal melanoma. Retina specialists abide by the Hippocratic oath to “first, do no harm,” but as surgeons, we inevitably harm tissue to benefit the patient, even while repairing damage. In the case of uveal melanoma, for medium and large tumors, very little doubt exists as to the imperatives of treatment. Even as care is individualized for each patient, the risk-to-benefit scale clearly weighs in favor of treatment. At Miami Ocular Oncology (MOOR), we collected preliminary data that show that tumor size impacts survival outcome, even when tumor genetics are poor (ie, class 2 GEP). This understanding has led us, and others, to advocate earlier treatment for uveal melanoma such that smaller tumors are now routinely being treated. Smaller tumors have also been shown to respond to nonradiotherapeutic treatments, including laser ablation.6 Laser ablation of small uveal melanoma was initially noted to have a failure of local tumor control in approximately 22% at 3 years. A second study by Mashayekhi et al reported an 11% local failure rate at 5 years and identified factors associated with a favorable tumor control rate.7 Unfortunately, anatomic complications and vision loss were not uncommon; approximately one-third of patients had a VA of 20/200 or worse after treatment.

Understanding the importance of tumor molecular genetics in uveal melanoma, we postulated that small tumors were more likely to have better genetics and that even in the face of tumor genetics associated with metastatic disease, earlier intervention could alter survival outcomes for these patients. At MOOR, we had previously established standard vitrectomy surgical approaches that utilized microincisional vitrectomy surgery (MIVS) employing valved trocars, controlled fluidics, membrane removal, and laser tumor ablation followed by suppression of postsurgical inflammation with intravitreal triamcinolone acetonide as a definitive treatment for small ocular melanoma. We offered this approach, coupled with tumor genetic determination with GEP classification beginning in 2005 as an alternative to radiotherapy in patients with small uveal melanoma. As a clinical treatment, the major focus was on informed patient consent, and we discussed both risks and benefits of this procedure with our patients. The first patients did well, with 1 patient exhibiting a class 1b tumor and the second exhibiting a class 2 GEP result. Using these data, we offered further treatment utilizing standard radiotherapy and referred our class 2 patient to a medical oncologist experienced in the use of GEP to potentially guide adjunctive systemic therapy and/or enhanced metastatic surveillance (another potential ethical dilemma). These first cases were presented at both national and international retina conferences.

With greater comfort in intraoperative tumor management coupled with excellent short-term patient results, we continued to offer this treatment approach to our small uveal melanoma patients. Informed consent discussions now included our understanding of specific risks and benefits to treatment. As our treatment numbers increased, we obtained Institutional Review Board (IRB) approval to retrospectively evaluate our patients to enable reporting beyond a small case series and to enable review for potential publication. With IRB approval, we reviewed our first patients treated with radiation-sparing primary MIVS-directed endolaser tumor ablation and presented these data at the International Society for Ocular Oncology. Initially, these data were reported coupled with a larger series of melanoma patients undergoing surgery. Ultimately, data on this unique series of 100 patients were presented at the Retina Society in 2015 and at additional meetings.8-11

Ideally, all surgeons continuously evaluate their surgical outcomes both on a case-by-case basis but also more globally. In the field of retinoblastoma, major treatment shifts have occurred in the absence of randomized clinical trial data. This requires the physician to maintain an active understanding of past, present, and potential future trends in treatment and consider their ethical responsibilities to patients when creating a treatment plan.


Ultimately, presentation of clinical data to a field of experts is required to achieve both feedback and dissemination of new ideas. Ideally, this occurs at a national or international meeting of experts. Finally, the existing “gold standard” is to publish the approach and its outcomes in the peer-reviewed literature. Physicians as a whole remain a skeptical group and the burden to new treatments is, and should remain, high. Nonetheless, if physicians do not move the field forward, patients will never reap the benefits of advances in treatment. Randomized clinical trials remain the ultimate bar for treatment evaluation, but in these rare and orphan diseases, clinical care by compassionate and caring physicians has defined virtually every advance in these 2 complex diseases.

All surgeons face ethical choices daily. How they make these choices defines them as clinicians and directs the evolution of the field. This article has highlighted the ethical dilemmas associated with treating patients with these relatively rare diseases. Retina specialists treating uveal melanoma practice in a unique situation in which their patients face their 2 greatest fears: blindness and cancer. Those treating ROP must consider the future quality of life of an infant with multiple comorbidities. Two final thoughts are important to consider: First, always use the patient’s best interest to guide any treatment choice, and second, remember that the field is fraught with disagreement about the nuances of care, but it behooves each retina specialist to listen to colleagues and be willing to understand that a different treatment approach may well be the right treatment approach. The goal is to see treatment through a lens focused on the best, ethical care of each patient. RP


  1. Gilbert C. Retinopathy of prematurity: a global perspective of the epidemics, population of babies at risk and implications for control. Early Hum Dev. 2008;84(2):77-82.
  2. Mintz-Hittner HA, Kennedy KA, Chuang AZ; BEAT-ROP Cooperative Group. Efficacy of intravitreal bevacizumab for stage 3+ retinopathy of prematurity. N Engl J Med. 2011;364(7):603-615.
  3. Mortality in patients with small choroidal melanoma: COMS report no. 4. Arch Ophthalmol. 1997;115(7):886-893.
  4. de Alba MA, Villegas VM, Gold AS, et al. Clinical findings and genetic expression profiling of three pigmented lesions of the optic nerve. Case Rep Ophthalmol Med. 2015;2015:590659.
  5. Gold AS, Murray TG, Markoe AM, et al. Uveal melanoma gene expression status post radiotherapy. Optom Vis Sci. 2014;91(1):e14-e17.
  6. Houston SK, Wykoff CC, Berrocal AM, Hess DJ, Murray TG. Lasers for the treatment of intraocular tumors. Lasers Med Sci. 2013 May;28(3):1025-1034.
  7. Mashayekhi A, Shields CL, Rishi P, et al. Primary transpupillary thermotherapy for choroidal melanoma in 391 cases: importance of risk factors in tumor control. Ophthalmology. 2015;122(3):600-609.
  8. Shah NV, Houston SK, Markoe A, Murray TG. Combination therapy with triamcinolone acetonide and bevacizumab for the treatment of severe radiation maculopathy in patients with posterior uveal melanoma. Clin Ophthalmol. 2013;7:1877-1882.
  9. Lonngi M, Houston SK, Murray TG, et al. Microincisional vitrectomy for retinal detachment in I-125 brachytherapy-treated patients with posterior uveal malignant melanoma. Clin Ophthalmol. 2013;7:427-435.
  10. Houston SK, Shah NV, Decatur C, et al. Intravitreal bevacizumab combined with plaque brachytherapy reduces melanoma tumor volume and enhances resolution of exudative detachment. Clin Ophthalmol. 2013;7:193-198.
  11. Shah NV, Houston SK, Markoe AM, Feuer W, Murray TG. Early SD-OCT diagnosis followed by prompt treatment of radiation maculopathy using intravitreal bevacizumab maintains functional visual acuity. Clin Ophthalmol. 2012;6:1739-1748.