Retinopathy of prematurity is a proliferating disorder of the immature retina that continues to be one of the leading causes of blindness in children of both the modern and developing world. Retinal detachment and macular dragging secondary to neovascularization can cause severe visual morbidity. Induced myopia and resultant amblyopia also represent significant causes of vision loss. 

The most recent guidelines recommend screening all babies born at 30 weeks or less gestational age, or those babies weighing less than 1,500 g.¹ Any baby that undergoes a difficult clinical course, as determined by the neonatologist, also meets criteria for screening. The initial exam is usually performed at four to six weeks of life and follow-up exams are repeated every one to two weeks until the retina is fully vascularized.¹

Current concepts of ROP describe it as a biphasic disease. Phase 1 consists of oxygen-induced vascular obliteration, followed by phase 2, which consists of hypoxia-induced vessel proliferation.^2,3 ROP is clinically described in terms of location (zone) and severity (stage).

Plus disease is posterior polar vascular dilation and tortuosity that represents increased blood flow. Its presence or absence is often a critical determination in deciding to treat or observe.

Neovascularization is mainly driven by vascular endothelial growth factor. Peripheral retinal ablation by conventional laser therapy is currently the gold standard of treatment and works by destroying peripheral retinal cells that produce VEGF. Laser treatment has a long history of effectiveness and safety. Cryotherapy is also an effective treatment and was used before development of laser. Scleral buckle and/or vitrectomy procedures are commonly used to repair Stage 4 or 5 ROP. Retinopathy of prematurity in Zone I is the most difficult to treat and has a high incidence of recurrence. The Early Treatment for Retinopathy of Prematurity (ET-ROP) study reported a better than 50 percent unfavorable outcome rate after laser photocoagulation in Zone I disease, if treated using traditional guidelines. Earlier treatment of high-risk, pre-threshold, Zone I babies still resulted in an approximately 30 percent unfavorable outcome rate.⁴ Unfavorable outcomes in this study were defined using both visual acuity, as determined by monocular grating with Teller cards, and structural criteria. After ROP has resolved, premature children should still have regular follow-up exams, as they are still at risk for strabismus, amblyopia and significant refractive error.⁴

Bevacizumab is a humanized anti-VEGF monoclonal antibody that has changed the treatment course of many retinopathies, including age-related macular degeneration and proliferative diabetic retinopathy. The role of anti-VEGF treatment for ROP is currently undefined. Several recent studies have investigated its place in preventing poor outcomes. The Efficacy of Intravitreal Bevacizumab for Stage 3+ Retinopathy of Prematurity (BEAT-ROP) study recently reported a significantly higher rate of recurrence with Zone I disease when treated with conventional laser therapy compared to intravitreal bevacizumab (42 percent vs. 6 percent, p=0.003). The authors conclude that intravitreal bevacizumab has increased efficacy over conventional laser therapy for stage 3+ retinopathy of prematurity within Zone I. They also note that there was continued vascularization of the peripheral retinal after intravitreal bevacizumab. Conventional laser therapy, in contrast, causes destruction of the peripheral retina and permanent visual field defects. During the study there were seven infants who died, five of whom were treated with bevacizumab and two with conventional laser.

Unfortunately, the study was not powered enough to adequately illustrate the drug’s safety. Systemic side effects of this drug in premature infants, who are still forming new blood vessels in many different organ systems, have not yet been determined. Time to recurrence of ROP was also different between the two groups. After conventional laser treatment, disease recurred 6.2 ±5.7 weeks after treatment. After intravitreal bevacizumab, disease recurred 16 ±4.6 weeks, with some infants having recurrent disease as late as 54 weeks. Since the study’s primary outcome end-age was 54 weeks, no recurrence of disease occurring after 54 weeks was recorded. The authors also changed the primary end-age halfway through the study.³

A retrospective, multicenter study based in Taiwan was recently conducted by Wei-Chi Wu, et al, where the medical records of patients with ROP who were treated with intravitreal bevacizumab were pooled for data analysis.5 The indications for treatment in this study met the criteria for type 1 ROP in the ET-ROP study, which includes Zone I disease of any stage with plus, Zone I stage 3 without plus, and Zone II stage 2 or 3 with plus. If the patients did not respond positively to the intravitreal injection within two to three weeks, conventional laser therapy was performed. For stage 4 or 5, an injection was used before surgery to reduce the chance of bleeding. A positive response was defined as disappearance of the tunica vasculosa lentis, dilation of pupils, disappearance of retinal vessel tortuosity and neovascularization, and improved vascularization of the peripheral retina.

Forty-nine eyes of 27 patients were included in the study. There were 41 eyes with stage 3 ROP, six eyes with stage 4A and two eyes with stage 5 ROP. Of the 41 with stage 3 ROP, nine eyes were Zone I ROP and 32 eyes were Zone II ROP. Thirty-six of these eyes received bevacizumab as primary treatment and five received bevacizumab as salvage treatment after showing no response from traditional laser therapy. In the 36 eyes that received bevacizumab as primary treatment, 32 (89 percent) showed ROP regression. Four eyes (11 percent) required supplemental laser treatment. In the five eyes that received bevacizumab as salvage therapy, all five (100 percent) had ROP regression. The percentage of Stage 3 ROP with or without plus disease was not noted in the study.⁵

In the eyes with Stage 4 ROP, two eyes showed regression without needing subsequent surgery and four eyes still required subsequent vitrectomy. In stage 5 ROP, the two eyes displayed decreased tortuosity after intravitreal bevacizumab, but surgical treatment was not successful in re-attaching the retina after several attempts. The authors conclude that bevacizumab as primary or salvage therapy caused neovascularization to regress and seemed to work best in stage 3 Zone I ROP. Although the study was not designed to compare the two treatments, the authors felt that for Zone I ROP, bevacizumab injection was more effective than conventional laser therapy. Ocular complications they report include transient retinal vessel sheathing and pre-retinal or vitreous hemorrhage. The authors state that the use of bevacizumab to treat ROP seems promising, but complications do occur and longer follow-up is necessary for these patients to prevent poor outcomes. This retrospective study is limited by lack of a control group and a lack of power. The authors of this study also report that large, randomized controlled trials are necessary to answer the many questions about intravitreal bevacizumab for ROP.⁵

A systematic literature review was recently published by Jonathan Micieli, and colleagues from McGill University.⁶ The authors searched OVID, Medline and Embase with subject headings of retinopathy of prematurity and angiogenesis inhibitors. Nine articles were selected. Six were case reports, two were retrospective reviews, and one was a prospective study. There were a total of 77 eyes from 48 infants. None of the studies were randomized trials.

The authors note that there is significant variability in the reported management of ROP with bevacizumab, including the dose and frequency, the long-term systemic effects, at what point to treat, and whether it should be used in conjunction with other therapies. The most common dose used was 0.75 mg, with a range of 0.4 mg to 1.25 mg used in the articles reviewed. None of the studies reported any systemic or local side effects; however, it is uncertain whether repeated doses, like treatment for macular degeneration, would still be safe.

The authors also note that there was a mixed outcome in patients with stage 4 disease. Of the 17 that were documented to have either Stage 4A or 4B ROP, who were treated initially only with bevacizumab, four eyes experienced resolution of the detachment without need for surgery, three eyes worsened, and the remaining 10 were treated with surgery. Use of intravitreal VEGF inhibitors in the presence of retinal traction can cause contraction and worsening of retinal detachment.

The authors urged that physicians who use bevacizumab in ROP do so with caution until further clinical studies are performed. Parents of children with ROP must be fully informed about the risks, benefits and alternatives if intravitreal bevacizumab is to be considered. If intravitreal bevacizumab is used to treat ROP, weekly follow-up for a prolonged period of time is necessary to detect recurrence or induced traction. The exact length of follow-up required has not yet been determined.⁶

Bevacizumab seems to have a promising future for the treatment of ROP; however, there is still a lot of uncertainty that surrounds its use. The potential for development of vascularized peripheral retina makes injection of VEGF inhibitors an exciting option. Bevacizumab injection for ROP also has potential for use in the Third World, where babies are screened through telemedicine and ophthalmologists are not locally available for laser treatment. ROP is likely to become more of a problem in these developing countries as neonatal care improves. The unpredictable post-injection course and the unknown systemic side effect profile in premature infants have created significant controversy and limited its widespread acceptance. More clinical research, especially randomized controlled trials, need to be performed before the safety, efficacy and guidelines for use are established.  REVIEW