Purpose
To report the outcomes of zone 1 aggressive posterior retinopathy of prematurity (ROP) treated with aggressive laser photocoagulation by a protocol-based approach over a 10-year period.
Design
Prospective, interventional case series.
Methods
setting: Tertiary eye institute and neonatal intensive care units in the twin cities of Hyderabad and Secunderabad in South India. study population: Babies screened under the Indian Twin Cities Retinopathy of Prematurity Screening (ITCROPS) program between January 1, 1997 and March 31, 2007. intervention: Multiple sessions of laser photocoagulation at any sign of plus disease with abnormal shunt vessels with or without neovascularization in zone 1. outcome measure: Based on the anatomic status of the retina at the final follow-up, the outcome was classified into good (completely regressed with no vitreoretinal changes), fair (regression with vitreoretinal distortion), or poor (progression into partial or total retinal detachment involving macula). Descriptive statistics (mean and range) were used for gestational age at birth, birth weight, and post-conceptional age at first presentation and at the first intervention. Trends in incidence of zone 1 ROP over 1 decade were analyzed. Outcomes were expressed in terms of proportions and 95% confidence interval (CI).
Results
Of the 3654 babies screened, 227 eyes of 115 babies were detected to have zone 1 aggressive disease, defined as zone 1 retinal vessels having abnormal closed-loop shunts, dilation and tortuosity, flat new vessels, or rapid progression. Laser was performed in 169 eyes. Disease regressed with good outcome in 142 eyes (84%) (95% CI, 77.6–89.2) and progressed to poor outcome in 13 eyes (7.69%) (95% CI, 4.1–12.8), while fair outcome was seen in 6 eyes (3.55%) (95% CI, 1.3–7.5). Thus 148 babies (87.5%) (95% CI, 81.6–92.14) had a favorable outcome. The outcomes in 8 treated eyes (4.7%) lost to follow-up were unknown.
Conclusions
Zone 1 aggressive ROP has a good outcome if the screening is done early; intervention is prompt and adequate, with frequent follow-up until complete regression is achieved.
Zone 1 retinopathy of prematurity (ROP) is known to rapidly progress to retinal detachment in the absence of intervention. Over the past 15 years, only a handful of studies have looked into the outcomes of zone 1 ROP when treated with laser photocoagulation. Most of these studies had small sample sizes and retrospective data and used varied criteria for intervention ( Table ).
| Year | No. of Eyes | Criteria for Intervention | Favorable Outcome | |
|---|---|---|---|---|
| Fleming and associates | 1992 | 10 | Plus disease | 100% |
| Capone and associates | 1993 | 30 | Threshold ROP | 83% |
| O’Keefe and associates | 1995 | 6 | Neovascularization | 100% |
| Vander and associates | 1997 | 19 | Earlier to threshold | 84% |
| 17 | Threshold | 82% | ||
| Katz and associates | 2000 | 22 | Plus disease with neovascularization | 60% |
| Axer-Siegel and associates | 2000 | 23 | Threshold | 85% |
| Shah and associates | 2005 | 72 | Fulminant posterior disease | 82% |
| Sanghi and associates | 2009 | 81 | Aggressive posterior ROP | 77% |
| Present series | 2010 | 169 | Plus disease and/or neoevascularization; abnormal looped shunt vessels | 87.5% (95% CI 81.6–92.14) |
The ITCROPS (Indian Twin Cities Retinopathy of Prematurity Screening) program was started 10 years ago, in 1997, in order to examine and analyze the computerized data pertaining to ROP in premature babies in the twin cities of Hyderabad and Secunderabad in the southern Indian state of Andhra Pradesh. Since then the data pertaining to the risk factors and the staging of and treatment outcomes of ROP have been prospectively collected in a custom-made computerized ROP database. Through March 2007 a total of 3654 neonates were examined under this program. The treatment outcomes in babies developing zone 1 ROP are presented in this report.
Methods
All evaluations and interventions were performed by trained retina specialists (S.J./A.H.) as per a previously published protocol using the indirect ophthalmoscope. Whenever possible, the retinal pictures were documented additionally using the Retcam digital camera (Clarity Medical Systems, Pleasanton, California, USA) or the video indirect ophthalmoscope (Appasamy, Chennai, India). Our screening protocol included screening of preterm infants weighing 1900 g or less and/or born before completion of 35 weeks gestational age (GA) within 20 to 30 days of birth. Babies were reviewed at twice-weekly intervals if no disease was seen until the retina was completely vascularized or non-vision-threatening ROP regressed completely. Detailed history included the birth weight, GA, and prenatal and postnatal problems and their management. The retinal findings were noted according to the International Classification of Retinopathy of Prematurity (ICROP) initially and later according to the revised classification.
Active zone 1 ROP (later renamed as aggressive posterior ROP [APROP]) was diagnosed if any 1 or more of the following criteria were fulfilled: the zone 1 retinal vessels did not show the typical dichotomous branching vascular pattern but showed abnormal closed-loop shunt vessels, there was dilation and tortuosity of vessels with or without flat new vessels in zone 1, and there was rapid progression of zone 1 disease within 4 to 7 days. All such cases were advised immediate laser ablation. Treatment was performed within 24 hours of diagnosing vision-threatening ROP in most cases and not later than 48 hours in any case. The eyes of the infants were reviewed every 3 to 7 days and additional laser performed to cover skip areas until the disease was completely arrested. Surgical treatment was considered in eyes developing progressive ridge elevation or tractional detachment of more than 4 clock hours.
All treatment was performed after obtaining written informed consent from the guardians. Laser photocoagulation was performed using topical anesthesia, either in the ophthalmologist’s operating room under the care of a pediatric anesthesiologist or in the neonatal intensive care unit. The pupils were dilated with tropicamide 1% eye drops and phenylephrine 2.5% eye drops instilled twice and not more than thrice starting 30 minutes before treatment. An 810-nm diode laser photocoagulator (Iris Medical OcuLight SLx, IRIDEX Corporation, Mountain View, California, USA) using the indirect ophthalmoscope and, in 16 eyes, the diopexy probe, was used. Treatment was directed to the avascular retina anterior to the ROP ridge. Confluent spots were placed using enough energy to produce light gray/cream-colored burns, avoiding intense white, full-thickness burns. Posterior areas were treated without scleral indentation, and higher energy was needed to treat these areas. More anterior areas were treated using gentle scleral indentation. A wire vectis was found convenient for this purpose. In all cases, the aim was to achieve complete regression by targeting the entire avascular retina as rapidly as possible. All cases had hemoglobin checked during the treatment and packed cell/blood transfusions were encouraged in consultation with the neonatologist in case hemoglobin was less than 10.0 gram percent.
After complete regression or complete retinal vascularization, follow-up was scheduled at 1, 3, 6, and 12 months and every year following that to assess visual acuity, refraction, and retinal status. Examination under short anesthesia was done whenever needed, especially to monitor intraocular pressure in surgical cases.
The primary outcome was the anatomic status of the retina at the final follow-up. Complete regression with good outcome was defined as an attached clinically normal-looking macula with no vitreoretinal traction and no vascular dilation, tortuosity, or new vessels anywhere in the retina at the last visit. Partial regression with fair outcome was defined as absence of plus disease, an attached macula with dragging of the disc and/or macula, or focal retinal detachment attributable to vitreoretinopathy but not involving the macula. Progression and poor outcome or unfavorable outcome was defined as development of either stage 4b or 5. A progressive 4a detachment extending circumferentially beyond 4 clock hours was categorized as progression for the study and was offered surgical treatment. Complete regression and partial regression were included under favorable outcome.
The data were collected and analyzed using Microsoft Excel 2003 (Microsoft Corp., Redmond, Washington, USA) and MedCalc statistical software (version 9.2.1.0; Mariakerke, Ghent, Belgium).
Results
Over a period of 10 years from January 1, 1997 to March 31, 2007, the ROP database recorded details of 3654 preterm babies. Of these, 227 eyes of 115 neonates presented with zone 1 ROP. The mean GA at birth was 29.63 weeks (range 24 to 36 weeks); the mean birth weight was 1228 g (range 580 g to 1900 g), and the mean post-conceptional age at first presentation was 33.9 weeks (range 28 to 52 weeks). The mean post-conceptional age at first intervention was 34.6 weeks (range 31 to 51 weeks). The mean follow-up from the time of the first examination was 10.34 months (range 1 month to 60 months). The trend of zone 1 ROP over the years is given in Figure 1 .
Since the aim of the current report was to study the effect of laser photocoagulation in preventing zone 1 eyes from progressing to retinal detachment, 13 eyes that presented to us with stage 4b zone 1 ROP were excluded from the final analysis.
Laser photocoagulation was performed in 169 eyes. In 142 eyes (84%) (95% confidence interval [CI] 77.6–89.2), ROP regressed, while ROP progressed in 13 eyes (7.69%) (95% CI 4.1–12.8). Six eyes (3.55%) (95% CI 1.3–7.5) showed partial regression. Thus 148 eyes (87.5%) (95% CI 81.6–92.14) had a favorable outcome. The outcomes in 8 eyes (4.7%) were not known because of loss to follow-up ( Figure 2 ). Of the 13 eyes that progressed even after laser, vitreoretinal surgery was done in 9 eyes, of which 7 had a favorable outcome with retinal reattachment, while in 2 eyes retina could not be attached.
