To determine whether birth weight less than 1500 g is a relevant guideline indicating the need for examination for retinopathy of prematurity (ROP) when gestational age at birth is 30 or more completed weeks.
A retrospective observational cohort study.
A total of 266 infants in a single institutional neonatal intensive care unit (NICU), whose gestational age at birth was 30 or more weeks but whose birth weight was less than 1500 g, were examined according to published guidelines. Infants with lethal congenital anomalies or major ocular abnormalities were excluded. Outcomes were vascularization in retinal zone III without a prior need for treatment, or ROP warranting treatment.
A study outcome was reached by 212 infants. Two hundred and eleven (99.5%) became vascularized through zone III without needing treatment. Only 1 (0.5%) required treatment for ROP. The 95% confidence interval for the occurrence rate of ROP requiring treatment in this cohort was 0.01%-2.60%.
Our results suggest that the occurrence rates of ROP requiring treatment in infants with gestational age 30 or more weeks and birth weight less than 1500 g is very low, and could indicate the need to revise examination guidelines for this subgroup of infants.
Current guidelines for examination of neonates who may require treatment for retinopathy of prematurity (ROP) have continued to evolve. While guidelines may vary worldwide, the criteria in the United States, and applicable to our neonatal intensive care unit (NICU) population during the period of this study, include gestational age 30 or fewer completed weeks or birth weight less than 1500 g.
We have observed that infants born at gestational age 30 weeks or later often have birth weights that call for examination under the guidelines, but only occasionally do they show even mild degrees of ROP. We undertook this study to determine the number of our highest-gestational-age neonates, considered to require examination for ROP under the criterion for birth weight, who developed ROP calling for intervention as recommended under the guidelines prevailing at the time.
Our results may form the basis for exploring, with more rigorous studies, whether existing guidelines should be reevaluated to allow fewer examinations of members of this high-gestational-age subgroup, without holding them exempt from all scrutiny.
Patients and Methods
The design and execution of this retrospective observational study was approved prior to initiation by the Icahn School of Medicine at Mount Sinai Program for the Protection of Human Subjects and conformed to the requirements of the United States Health Insurance Portability and Privacy Act and the Declaration of Helsinki.
We reviewed the records of infants born during the period January 1, 2001 through December 31, 2010 who were cared for in our single-center institutional level IIIC NICU. As is standard in neonatology practice and research, we ascertained gestational age by best obstetrical estimate, a combination of menstrual dating and first trimester sonography; this was corroborated by physical examination at birth. Fractional days beyond the estimated number of completed weeks of gestation were not “rounded up.” Exclusion criteria were those of the CRYO-ROP and ET-ROP studies, that is, infants with lethal congenital anomalies or major ocular abnormalities in 1 or both eyes. We recorded the postmenstrual age, in completed weeks, and results of each examination on all infants, without considering racial groups separately.
Examination was performed after topical administration of a cycloplegic/mydriatic agent by binocular indirect ophthalmoscopy and scleral indentation. A fellowship-trained pediatric ophthalmologist (E.L.R.) performed approximately 85% of the examinations over the entire 10-year period. The remaining examinations were provided by a vitreoretinal surgeon (R.N.G.) and a subspecialist in medical retinal disorders (S.E.B.), both having broad experience with childhood ophthalmologic conditions within their fields of interest and with ROP examination.
After initial evaluation at 4-5 weeks of chronological age, examinations were performed at 2-week intervals if no ROP was present and at 1-week intervals if any degree of ROP was detected and had not yet resolved spontaneously. Infants requiring ophthalmologic follow-up after discharge from the NICU received those examinations in our outpatient facility.
Our outcome measures were the occurrence of either ROP warranting treatment or confirmed retinal vascularization present in zone III, with or without prior spontaneously resolved milder ROP. We discontinued follow-up when 1 of these outcomes was present. We assumed that eyes once reaching the need for treatment would not later show spontaneous regression to the point of no detection, so that the absence of examinations between those at planned intervals would not have revealed severe ROP.
Statistical comparisons were made with Fisher exact test or Student t test (2-tailed) as appropriate. Our chosen level for statistical significance was P < .05. We ascertained the 95% confidence interval for our observed occurrence rate of severe ROP from the analysis-one proportion module of the NCSS statistical software model, version 9.0 (NCSS, Kaysville, Utah, USA).
Of 780 neonates examined for ROP during our 10-year study period, there were 332 who met study gestational age and birth weight criteria. Nine of these died or were transferred for care at other institutions prior to any examination. An additional 57 eligible infants did not undergo ophthalmologic examinations at our institution.
Complete medical records including ophthalmologic examination history were available for 266 infants. There were 128 male and 138 female infants; this is not significantly different ( P = .73) from an equal sex distribution. Their gestational ages at birth are listed in Table 1 and their birth weights in Table 2 .
|Gestational Age (Wk)||No.of Patients|
|Birth Weight (g)||No. of Patients||Patients With ROP|
One of our 2 defined study outcomes was reached by 212 infants. Only 1 (0.5%) required treatment for ROP. This infant was born at 32 weeks of gestation weighing 905 g. He was transferred to our NICU for unrelated surgical care and did not show any ROP on examination at postmenstrual age 34 weeks. His diagnosis of ROP and need for treatment was reached following transfer back to the original facility. Because he received care and both pre- and posttreatment ophthalmologic examinations at our institution, he is included in our cohort.
Two hundred and eleven babies (99.5%) were observed to become vascularized in zone III. The postmenstrual ages at which this was first noted are shown in Table 3 . Of these, 168 of the 211 (79.6%) reached zone 3 at or prior to 38 weeks postmenstrual age.