Near Confluent Laser Photocoagulation for the Treatment of Threshold Retinopathy of Prematurity FREE

Retinopathy of prematurity (ROP) is a proliferative disease of the developing retinal vasculature in premature infants, which may lead to severe visual loss.¹ The Multicenter Trial of Cryotherapy for Retinopathy of Prematurity showed that cryoablation of the avascular peripheral retina significantly reduces the progression of threshold ROP.^{2 - 6} Threshold ROP is defined as stage 3 disease in zones 1 or 2 with at least 5 contiguous or 8 cumulative clock hours of extraretinal fibrovascular proliferation in the presence of plus disease.^{2 - 7} It was subsequently demonstrated that indirect laser photocoagulation is at least as effective as cryotherapy in preventing unfavorable outcomes of ROP.^{8 - 25} Indirect laser photocoagulation permits a precise and relatively atraumatic treatment to the retina, is more convenient and technically easier to administer, and results in less ocular and systemic adverse effects.

The wavelength and pattern of laser application in ROP may affect the outcome. Comparisons have been made between argon and diode lasers, a dense vs less dense laser pattern, and treatment to the avascular peripheral retina with and without treatment of the ridge.^{14 ,26 - 29} Since 1997, one of us (D.E.) has been using the diode laser indirect ophthalmoscope to deliver a near confluent laser pattern to the peripheral avascular retina between the ridge and the ora serrata. To further evaluate the effect of the density of laser photocoagulation on the progression of threshold ROP, we compared a near confluent pattern of indirect infrared diode laser photocoagulation (wavelength, 810 nm) in threshold ROP with the previously reported patterns of laser treatment.^{12 ,26 ,30} Treatment was applied to the peripheral avascular retina with a near confluent pattern of laser spots, spaced approximately 0.25 burn width apart. By the end of the treatment session, the laser burns had expanded to a confluent pattern.

A retrospective medical record review was performed on premature infants who underwent treatment for threshold ROP from 1997 to 2002 at Kresge Eye Institute (Detroit, Mich), Hutzel Hospital (Detroit), Children’s Hospital of Michigan (Detroit), Huron Valley Hospital (Commerce Township, Mich), and Detroit Riverview Hospital. All of the patients underwent preoperative examination and treatment by 1 surgeon (D.E.). A total of 58 eyes from 31 consecutive patients were diagnosed as having threshold ROP and were treated within 48 hours of diagnosis. Fourteen eyes of 8 patients with zone 2 disease were excluded from the study: 8 eyes of 4 patients had only 1 month of follow-up after laser treatment, 1 patient died 1 month after laser treatment, 2 eyes of 1 patient were treated with cryopexy owing to small pupils and extensive tunica vasculosa lentis, and 2 eyes of the remaining 2 patients had received previous laser treatment at another institution and received supplemental laser treatment by us. (In all 14 eyes, the retinopathy had regressed at the last examination.) Ten eyes of 5 patients with zone 1 disease and 34 eyes of 18 patients with zone 2 disease were included in the study. All of the eyes were treated with indirect diode laser photocoagulation and had at least 3 months of follow-up.

The mean gestational age of the infants was 24.7 weeks (range, 23-27 weeks), and the mean birth weight was 674 g (range, 520-936 g). All eyes had reached a minimum of 5 contiguous or 8 total clock hours of extraretinal fibrovascular proliferation with vascular dilation and tortuosity in the posterior pole (plus disease).

A complete examination was performed, including indirect ophthalmoscopy with 360° of scleral depression. Pupillary dilation was achieved with 2.5% phenylephrine hydrochloride and 1% tropicamide (1 drop of each, repeated once after 5 minutes). Examination findings were recorded on standardized data sheets. The zones and stages were documented as described by the Committee for the Classification of Retinopathy of Prematurity.⁷ Written informed consent from a legal guardian was obtained before treatment. If both eyes required treatment, they were treated at the same session.

A portable indirect infrared diode laser (IRIS Medical Inc, Mountainview, Calif), a lid speculum, a scleral depressor, and a 28-diopter aspheric lens (Volk, Mentor, Ohio) were used for all of the treatments. One drop of proparacaine was placed in each eye before treatment. Balanced salt solution drops provided a clear view for the duration of the procedure. In all cases, the laser treatment was applied to the avascular retina between the anterior edge of the fibrovascular ridge and the ora serrata for 360°. Areas of retina that were covered with preretinal hemorrhage were not treated. A moderate white burn was the target intensity that was achieved at a power range of 300 to 500 mW. Duration of a single spot was 100 milliseconds. The laser burns were spaced approximately 0.25 burn width apart. By the end of the treatment session, the laser burns had expanded to a confluent pattern.

Follow-up examinations were performed at weekly intervals until complete regression of extraretinal fibrovascular tissue had taken place. Once complete regression was achieved, follow-up was extended to biweekly and then monthly intervals until 6 months of age. Progression was defined as the development of any retinal detachment (stages 4A, 4B, and 5).

Ten eyes of 5 patients with zone 1 disease were included in the study (Table 1). Three of the treated infants were male, and 3 were black. The mean gestational age of the infants was 24.6 weeks (range, 24-26 weeks), and the mean birth weight was 679 g (range, 620-770 g). The mean number of clock hours of stage 3 ROP was 10.6 (range, 8-12). Treatment intensity ranged from 300 to 500 mW, with a mean ± SD number of spots of 2534 ± 455 (range, 2100-3378). These infants were followed for up to 40 months. Both eyes of 1 infant progressed to stage 5, which required vitrectomy surgery. One eye of 1 infant developed anterior-posterior traction on the macula and underwent lens-sparing vitrectomy with a good final anatomic outcome. Neovascularization in the remaining 7 eyes with zone 1 disease regressed after laser treatment, and none of these eyes required additional treatment. Seven (70%) of 10 eyes with zone 1 disease had a favorable outcome (95% confidence interval [CI], 42%-98%).

Thirty-four eyes of 18 infants with zone 2 threshold ROP were included in the study (Table 2). Eight of the treated infants were male, and 8 were black. The mean gestational age was 24.7 weeks (range, 23-27 weeks), and the mean birth weight was 673 g (range, 520-936 g). The mean number of clock hours of stage 3 ROP was 8.4 (range, 5-12). Treatment intensity ranged from 300 to 500 mW, with a mean ± SD number of spots of 1850 ± 487 (range, 1030-2689). None of the eyes required supplemental laser treatment. Mean follow-up was 14.9 months. All eyes with zone 2 disease (100%) had a favorable outcome (95% CI, 90%-100%).

No cases of iris burn, cataract formation, or anterior segment ischemia in any of the treated eyes (both zone 1 and 2) were noted during the follow-up visits. In some infants, bradycardia and/or oxygen desaturation occurred during the laser treatment, which resolved after temporary discontinuation of the treatment.

Retinopathy of prematurity is a major cause of blindness in newborns.^{31 - 32} Ideally, premature infants are examined according to screening guidelines established by the American Academies of Ophthalmology and Pediatrics and by the Association for Pediatric Ophthalmology and Strabismus, and treatment is initiated shortly after threshold disease is reached. Recently, the Early Treatment of Retinopathy of Prematurity Cooperative Group recommended peripheral retinal ablation treatment to be considered for zone 1, any stage ROP with plus disease; zone 1, stage 3 ROP with or without plus disease; and zone 2, stage 2 or 3 ROP with plus disease.³³ The main goal of treatment is to prevent the progression to retinal detachment, because the treatment of advanced stages of ROP (stages 4 and 5) is associated with a poor functional outcome.^{34 - 36}

Although the exact cause of ROP is not known, it develops in premature infants with low birth weight.³⁷ The disease is characterized by proliferation of abnormal fibrovascular tissue at the border of vascularized and nonvascularized retina.³⁸ The liberation of various angiogenic molecules by cells in the hypoxic retina may contribute to the progression of the disease.³⁹ It is likely that vascular endothelial growth factor is one of the molecules that contributes to vascular proliferation and vasodilation.^{39 - 44} The production of vascular endothelial growth factor is regulated by hypoxia, and both retinal pigment epithelial cells and glial cells contribute to the expression of vascular endothelial growth factor in the presence of hypoxia.^{45 - 47}

As has been extensively reported, the ablation of the hypoxic peripheral retina with cryotherapy or photocoagulation increases the likelihood of regression of the disease.^{2 - 26} It was shown that a dense pattern of laser photocoagulation (649 spots for zone 2 disease and 1253 spots for zone 1) was more successful than a less dense pattern (457 spots for zone 2 disease and 592 spots for zone 1) in reducing disease progression.²⁶ One may have expected this outcome, since more thorough ablation of hypoxic tissue would result in less angiogenic factors. These findings suggest that a near confluent laser pattern may be even more successful than the reported dense treatment pattern. Because the small sample size of our patients with zone 1 disease (10 eyes of 5 infants) would not enable us to directly compare our results with the recently reported data for dense and less dense laser photocoagulation patterns, we compared our results for zone 2 retinopathy.²⁶ Our mean ± SD number of laser spots, 1850 ± 487, was much higher than that in the reported study for dense treatment of zone 2 eyes (649 spots) and less dense treatment of zone 2 eyes (457 spots). None of our patients progressed to stage 4A, 4B, or 5. The rate of progression in our group (0%) was lower than the reported study with dense laser treatment of zone 2 eyes (3.8%) and the less dense laser pattern (21.2%). The rate of re-treatment in our group (0%) was also lower than the re-treatment rates for the reported dense laser treatment (37.5%) and the less dense laser pattern (35.3%); however, the reported re-treatment rates involved patients with both zone 1 and zone 2 disease (our data include only zone 2; however, none of our zone 1 patients required re-treatment).

In a recent retrospective study,²⁹ however, it was shown that although confluent laser photoablation lowered the rate of supplemental treatment, the overall progression to stage 4 or 5 was similar to other treatment patterns. The number of laser spots in the eyes receiving confluent laser treatment (zone 1 or zone 2) ranged from 693 to 4535 spots, with a mean ± SD of 1943 ± 912 spots.²⁹ A direct comparison of our results with this report is not possible, since the authors divided zone 2 into anterior and posterior and grouped their data for zone 1 and posterior zone 2 together. This division may be useful because posterior zone 2 disease may respond differently from anterior zone 2 disease; however, the retrospective nature of our study did not enable this division.

In our study, 10 eyes of 5 patients with zone 1 disease and 34 eyes of 18 patients with zone 2 threshold ROP were treated with near confluent laser therapy within 48 hours of diagnosis. Although the number of laser spots is a useful measure of the extent of treatment, it is less important than the surface area of the treated retina, since spot size may vary when using the indirect ophthalmoscope to deliver laser photocoagulation. Spot size depends on several variables, including distance of the handheld aspheric lens from the eye, dioptric power of the aspheric lens, burn duration, and burn intensity. In addition, the visible burn expands shortly after treatment. Our goal was to create burns that were initially approximately 0.25 burn width apart, which expanded to a near confluent pattern by the end of the treatment session (typically 30 to 40 minutes per eye). In this manner, we attempted to treat almost the full extent of the hypoxic retinal surface area.

We acknowledge that others may already have treated the avascular area with a large number of laser spots.^{8 - 9 ,12 ,15 ,18 - 19 ,21 ,26 ,29 ,48 - 52} The number of reported laser spots ranges from approximately 300 to 4535 per eye. There is great variability in the number of spots within each study, leading to a mean number of laser spots that is generally lower than ours. Furthermore, we emphasize that the treatment of the entire area of the avascular retina is more important than the actual number of spots.

The most important outcome in assessing the treatment of threshold ROP is the progression of the disease to advanced stages. After a mean follow-up of 14.9 months (for zone 2 only), none of the eyes in our zone 2 series progressed to stage 4 or 5. One may hypothesize that the lower rate of progression for our near confluent treatment pattern for zone 2 disease (0%) compared with the reported dense treatment pattern for zone 2 disease (3.8%) and less dense pattern for zone 2 disease (21.2%) is based on more thorough ablation of the hypoxic tissue (in the near confluent treatment) that is responsible for angiogenesis.

None of the eyes in our series for zone 2 disease required re-treatment. If progression of neovascularization occurred despite a near confluent laser pattern, there would be essentially no untreated peripheral avascular retina left for re-treatment (except for patients who experienced preretinal hemorrhage in the avascular peripheral retina before initial treatment). Our re-treatment rate for zone 2 disease (0%) compares favorably with the reported re-treatment rate in the dense treatment pattern (37.5%) and less dense pattern (35.3%).²⁶ A statistical comparison is not possible, since the reported re-treatment rates involved patients with both zone 1 and zone 2 disease and our sample size for zone 1 disease was too small for direct comparison (although none of our zone 1 patients required re-treatment).

It has been suggested that potential adverse effects of intense laser photocoagulation of the avascular retina include anterior segment ischemia and necrosis.^{48 ,53} It is not known whether this is the result of intense ablation of the retina or of the inadvertent treatment of anterior segment structures. In none of our infants did we observe signs of anterior segment ischemia. The development of cataract is another reported adverse effect of transpupillary laser treatment.^{11 ,48 ,53 - 61} The incidence of cataract formation is less with infrared diode lasers compared with argon lasers. In our series, we did not observe the development of cataract in any of our patients during the follow-up period. Although the exact mechanism is not known, it has been hypothesized that the laser energy may be absorbed by the dense tunica vasculosa lentis (when present), or it may lead to the destruction of long posterior ciliary arteries along the 3- and 9-o’clock meridians, leading to anterior segment ischemia and/or cataracts.⁵⁴

Race may play a role in the progression of ROP in low-birth-weight infants.⁶² Severe, vision-threatening ROP occurs with greater frequency in white infants than in black infants. In our study, 8 of 18 patients with zone 2 disease were black. This may partially explain our good outcome in this group of patients.

Although our study had a limited number of patients with zone 1 threshold ROP, our results for zone 2 threshold ROP suggest that, at least in this group of eyes, a near confluent laser treatment pattern may be associated with a lower incidence of progression to higher stages of the disease and a lower re-treatment rate.

Correspondence: Dean Eliott, MD, Kresge Eye Institute, Wayne State University School of Medicine, 4717 St Antoine, Detroit, MI 48201 (deliott@med.wayne.edu).

Submitted for Publication: February 26, 2004; final revision received July 28, 2004; accepted August 6, 2004.

Financial Disclosure: None.

Funding/Support: This study was supported in part by the Ronald G. Michels Fellowship Foundation (Baltimore, Md) and Research to Prevent Blindness, Inc (New York, NY).