Chicago—A birthweight of less than or equal to 1,250 g was a dependable criterion for screening of retinopathy of prematurity (ROP) in a small, single-center retrospective study. Using this standard also reduced the number of screenings performed and consequently lowered the cost of screening, said Shailesh K. Gupta, MD, assistant professor of ophthalmology, University of Florida College of Medicine, Jacksonville.
Speaking at the American Academy of Ophthalmology meeting, Dr. Gupta stated that the purpose of the study was to determine whether birthweight of 1,250 g could provide a safe and efficient cutoff point to detect high-risk ROP and to limit the number of screenings without missing cases of ROP.
"Published studies indicate that using a criterion of ≤1,250 g may be reasonable," Dr. Gupta said. "If you look through all the data, what they show is that with the exception of one or two infants, nothing was missed greater than stage 3." The studies he referred to included infants of <1,250 g or, in a few instances, <1,200 g.
Retrospective chart review
The study that Dr. Gupta, Dr. K.V. Chalam, Dr. Selina Lin, and Lorene Ward performed was a retrospective chart review of 208 infants screened for ROP in a university setting over a 2-year period. Once the babies were screened and followed, the data were divided into two categories: Group 1, babies with birthweight ≤1,250 g, and Group 2, those more than 1,250 g.
The initial examination was performed at 6 weeks chronological age or 32 weeks gestational age, whichever was earlier. Examinations were performed weekly, or more often if suggested by the severity of disease. Laser treatment was performed within 72 hours when indicated. Eyes were staged according to the international classification for ROP.
Data collected included maximum stage of ROP achieved in each eye, evidence of PLUS disease, treatment if needed, birthweight, gestational age, race, and gender. There were 158 patients in Group 1, including 85 males and 73 females, and 50 patients in Group 2, which consisted of 31 males and 19 females. In Group 1, 93 patients were black and 63 were white. In Group 2, 25 were black and 23 were white. Two patients were of other races in each group. The mean gestational age was 26.8 weeks for Group 1 and 30 weeks for Group 2.
Dr. Gupta found that 116 eyes in Group 1 had ROP; 54 eyes had a maximum stage of 1, 35 eyes had a maximum stage of 2, 24 eyes a maximum stage of 3, two eyes a maximum stage of 4, and one eye, a maximum stage of 5. Twenty-six eyes were noted to have PLUS disease, and 24 were treated.
In Group 2, one eye in one patient had ROP diagnosed, which did not exceed stage 1.
Investigators then calculated sensitivity, specificity, and positive predictive value (PPV) using current guidelines and a birthweight/gestational age index calculated by multiplying birthweight by gestational age and dividing by 1,000. Sensitivity and specificity were 91% and 40%, respectively, and PPV was 37%.
"When using a singular criterion of just 1,250 g, there's a sensitivity of 98% and a specificity of 32%," Dr. Gupta said. "If you use a single criterion of 1,500 g, sensitivity is a little bit more, but your specificity drops down to 0.70%, and predictive value is 27%."
The researchers also plotted data comparing gestational age with birthweight with a cutoff point of 1,250 g. Most data points indicating ROP were clustered at the lower gestational ages and birthweights, with only one above 1,250 g.