Objective. To determine appropriate upper limits for gestational age and birth weight when screening infants for retinopathy of prematurity (ROP).
Design. Retrospective survey.
Setting. Tertiary neonatal intensive care nursery.
Patients. Seven hundred seven infants born July 1, 1990 to June 30, 1996 and screened for ROP according to the 1988 to 1996 American Academy of Pediatrics guidelines.
Outcome Measures. Maximum stage of ROP with respect to birth weight and gestational age.
Results. No ROP more than Stage 1 was observed in infants with gestational ages ≥32 weeks or birth weights ≥1500 g. All cases of threshold and Stage 4 ROP were confined to infants with gestational ages ≤30 weeks or birth weights <1200 g.
Conclusions. The latest American Academy of Pediatrics screening guidelines for ROP are discretionary for infants with birth weights >1500 g or gestational ages >28 weeks. If ROP screening is limited to infants with birth weights of ≤1500 g, 34.2% fewer infants would require screening compared with the previous <1800 g recommendation, while missing no cases of ROP more than Stage 1. A gestational age cut-off of ≤28 weeks, however, is less desirable, and could potentially miss several infants with more advanced retinopathy (including Stage 4). If ROP screening criteria were instead modified to include infants of gestational ages <32 weeks, the number of patients requiring screening could be reduced 29.1% compared with the previous recommendation of <35 weeks, again without missing any cases of ROP more than Stage 1. Use of such a screening strategy (birth weight <1500 g or gestational age <32 weeks) is predicted to save in excess of 1.5 million dollars annually in the United States, while missing no cases of ROP more than Stage 1.
- AAP =
- American Academy of Pediatrics •
- ROP =
- retinopathy of prematurity •
- CI =
- confidence interval
The American Academy of Pediatrics (AAP), the American Association for Pediatric Ophthalmology and Strabismus, and the American Academy of Ophthalmology together have recently recommended new screening guidelines for the timely detection of retinopathy of prematurity (ROP) in prematurely born infants.1Whereas these new guidelines mandate screening for infants with birth weights of ≤1500 g or gestational ages of ≤28 weeks, the screening of larger infants is left to the discretion of the attending pediatrician or neonatologist. Although the Multicenter Trial of Cryotherapy for ROP2 greatly expanded our knowledge of the incidence and natural history of ROP in infants with birth weights <1250 g, current incidence data for ROP in infants with birth weights >1500 g are sorely lacking, raising the question as to which of these larger infants should be screened under the new AAP guidelines. With few exceptions,3-5 most recent reports providing ROP incidence data have included only patients with birth weights ≤1500 g.6-14 In hopes that the decision of which infants to screen with a birth weight of >1500 g or a gestational age of >28 weeks can become less discretionary and more evidence-based, we wish to report incidence figures for ROP derived from a comprehensive screening program based on the 1988 to 1996 AAP guidelines.15
We performed a retrospective review of all 3908 infants admitted to the Intensive Care Nursery of the University of Tennessee Medical Center at Knoxville throughout a 6-year period dating from July 1, 1990 to June 30, 1996. Of these 3908 infants, 793 were eligible for ROP screening under the former AAP guidelines by virtue of: 1) birth weight <1300 g or gestational age of <30 weeks irrespective of oxygen exposure; or 2) birth weight <1800 g or gestational age <35 weeks with exposure to supplemental oxygen.15 Gestational age determinations were made using a best obstetric estimate including first or second trimester ultrasounds combined with menstrual dating when available. In the absence of reliable obstetric dating, gestational ages were estimated using the New Ballard Score.16 Of the 793 infants eligible for screening, 82 infants expired before their first examination. Of the remaining 711 infants, only 4 (0.6%; birth weight range, 1260 to 1760 g; gestational age range, 32 to 34 weeks) were transported at 4 to 31 days postnatal age to another hospital before screening, and were excluded from analysis. (The smallest of these infants was ultimately found to have no ROP, the largest infant was not screened, and no information was available on the remaining 2 infants.) For each of the remaining 707 infants, dilated funduscopic exams were performed by one of three experienced retinologists starting at 5 to 7 weeks postnatal age or hospital discharge, whichever came first. Infants were screened serially at intervals determined by the retinologist until either retinal vascularization was complete or the retinopathy had regressed. The maximal stage of retinopathy observed was recorded for each infant according to the International Classification of Retinopathy of Prematurity.17 The incidence and severity of ROP were then examined with reference to gestational age (in weeks) and birth weight range (in 100-g increments).
Overall screening results and demographic data are shown in Table1. In more than half of the infants screened [54.3%; 95% confidence interval (CI), 50.6 to 58.0%], no ROP was detected. One hundred eighty-two infants (25.7%; 95% CI, 22.5% to 29%) had Stage 1 ROP, 74 (10.5%; 95% CI, 8.2% to 12.7%) had Stage 2, 35 (4.9%; 95% CI, 3.4% to 6.5%) had subthreshold Stage 3, 28 (4.0%; 95% CI, 2.5% to 5.4%) had threshold Stage 3, and 4 (0.6%; 95% CI, 0.0% to 1.1%) had Stage 4 ROP, respectively. Although 8 of the 707 study infants expired before hospital discharge (3 with no ROP, 3 with Stage 1, and 1 each with threshold Stage 3 and Stage 4 ROP, at mean postmenstrual ages of 40.7, 47.2, 68.3, and 64.3 weeks, respectively), in only one instance (no ROP, 24 week gestation, postmenstrual age at death 31.4 weeks) did death occur before full retinal vascularization or ROP regression.
The maximum stages of ROP, classified by gestational age and birth weight ranges, are depicted in Figs 1 and2, respectively. No ROP more than Stage 1 was observed in infants with gestational ages ≥32 weeks or birth weights ≥1500 g. All cases of threshold and Stage 4 ROP were confined to infants with gestational ages ≤30 weeks or birth weights <1200 g.
Compared with the previous AAP birth weight screening limit of <1800 g, use of screening limits of <1700 g, <1600 g, or <1500 g would effectively lower the number of infants requiring ROP screening by 19.8% (95% CI, 16.9% to 22.7%), 26.9% (95% CI, 23.6% to 30.1%), and 33.9% (95% CI, 30.5% to 37.4%), respectively, with no apparent loss of sensitivity for ROP more than Stage 1. Similarly, compared with the previous AAP gestational age limit of <35 weeks, use of screening limits of <34 weeks, <33 weeks, or <32 weeks would reduce the number of infants requiring examination by 8.2% (95% CI, 6.2% to 10.2%), 16.1% (95% CI, 13.4% to 18.8%), and 29.1% (95% CI, 25.8% to 32.5%), respectively, again seeming to miss no cases of ROP more than Stage 1. However, use of the newly proposed gestational age cutoff of <28 weeks would have the undesirable consequence of missing several cases of ROP more than Stage 1, namely 10 of 74 infants with Stage 2 disease (13.5%; 95% CI, 6.9% to 23.6%), 5 of 63 infants with Stage 3 disease (7.9%; 95% CI, 1.4% to 17.5%), and 1 of 4 infants with Stage 4 disease (25%; 95% CI, 0.6% to 80.6%).
In today's increasingly cost-conscious health care climate, the impetus is to utilize cost-effective, evidence-based strategies aimed at disease prevention, detection, and/or treatment. With the demonstration that cryotherapy could significantly reduce the likelihood of unfavorable visual outcomes in infants with threshold ROP,2 the importance of an effective screening strategy for at-risk infants became readily apparent. Whereas Javitt et al18 concluded that ROP screening and cryotherapy in infants with birth weights <1250 g could be cost-effective, similar screening strategies, aimed at larger, more mature infants (whose risk for threshold ROP is considerably less), may be quite cost-ineffective. The current AAP guidelines recommend ROP screening for all infants with birth weights ≤1500 g or gestational ages ≤28 weeks, plus any larger infants with unstable clinical courses felt to be at high risk for ROP by their attending physicians.1 The published data cited to support these new recommendations,2 19 20 however, include only those infants enrolled in the CRYO-ROP trial, all of whom had birth weights ≤1250 g. The current risk for ROP in larger, more mature infants is much less well documented. In our literature review, we were unable to identify any studies that utilized the former AAP screening criteria from which their validity could be tested.
Having followed the former AAP guidelines for ROP screening for the past several years, we found that 54.3% of the infants formerly targeted for screening failed to show any ROP whatsoever. Extrapolating from the data of Javitt et al,18 and assuming an annual United States birth cohort of 60 000 premature infants with birth weights 500 to 1800 g, at a conservative $84 screening cost per infant, approximately $1.5 million to $1.7 million could be saved annually by lowering ROP screening criteria to a gestational age of <32 weeks or a birth weight of <1500 g, respectively.
A limitation of the current study is that our cohort is from but a single institution and our sample size is relatively small compared with the 4099 infants enrolled in the CRYO-ROP study.2 A larger sample size might yield different conclusions, and our reported incidence of ROP may not be similar to that found in other institutions. As to the potential applicability of our results to other tertiary nurseries, we wish to offer two comparisons: 1) if only infants with birth weights <1250 g from our nursery are considered (N = 343), the proportion of infants with Stage 2 ROP (21%) and Stage 3 or more ROP (19%) is reassuringly similar to that reported in the multicenter CRYO-ROP study (21.9% Stage 2 and 18.4% more than or equal to Stage 3), and 2) compared with our fellow Vermont-Oxford nurseries reporting data on infants with birth weights 501 to 1500 g born 1992 to 1996, our nursery rankings for percent mortality and percent Stage 3 ROP were in the 8th to 15th and 74th to 85th percentiles, respectively.
Our screening criteria recommendations are based, in part, on the observation of no threshold ROP in either the 294 infants with gestational ages >30 weeks or the 399 infants with birth weights ≥1200 g. With our current sample size, we can estimate, with 95% confidence, that the true incidence of threshold disease for infants with gestational ages >30 weeks or birth weights ≥1200 g is no higher than 0.9%.21
More to the point, however, is an estimate of the risk of threshold disease in those larger premature infants who would no longer be screened under our recommended guidelines, namely those infants with birth weights 1500 to 1800 g or gestational ages 32 to 35 weeks. This situation is depicted graphically in Fig3, where cases of threshold ROP are plotted by birth weight and gestational age. Assuming a bivariate normal distribution for gestational age and birth weight, the bootstrap technique22 provides such an estimate, namely 4.5 cases per 100 million infants (95% CI, 1.11 E16 to 1.77 E5). Expressed differently, and assuming a yearly cohort of 20 000 eligible premature infants with birth weights 1500 to 1800 g, our recommended screening strategy would be expected to miss no more than one case of threshold ROP every 1111 years (95% CI, 2.94 years to 4.5 E13 years).
In summary, ROP, with very few exceptions,4 23 is presently a disease of very premature, low birth weight infants. Current ROP screening guidelines are discretionary for infants with birth weights >1500 g or gestational ages >28 weeks. Our data suggest that a birth weight cut-off of ≤1500 g seems unlikely to miss any cases of ROP more than Stage 1. Use of this weight limit would reduce the number of infants screened under the previous guidelines by 34.2%. However, use of a gestational age limit of ≤28 weeks may be overly restrictive, potentially missing several cases of ROP Stage 2 or more. A gestational age limit of ≤32 weeks would reduce the number of infants screened by 29.1% compared with the previous AAP guidelines, again missing no cases of ROP more than Stage 1. These criteria (birth weight ≤1500 g and gestational age ≤32 weeks) are identical with those recently proposed in the United Kingdom by the Joint Working Party of the Royal College of Ophthalmologists and the British Association of Perinatal Medicine.24 Use of such a screening strategy in the United States is predicted to save in excess of 1.5 million dollars annually, while missing no cases of ROP more than Stage 1, a disease typically associated with minimal long-term morbidity.25
The authors gratefully acknowledge the tireless support of Retina Associates of Knoxville PC (John C. Hoskins, MD; Joseph M. Googe Jr, MD; and James H. Miller Jr, MD).
- Received June 9, 1997.
- Accepted December 16, 1997.
Reprint requests to (K.W.) University of Tennessee Medical Center, Department of Pediatrics, U-113, 1924 Alcoa Hwy, Knoxville, TN 37920.
- Joint Statement of the American Academy of Pediatrics, the American Association for Pediatric Ophthalmology and Strabismus, and the American Academy of Ophthalmology
- Jandeck C,
- Kellner U,
- Kössel H,
- Bartsch M,
- Versmold HT,
- Foerster MH
- Fielder AR,
- Robinson J,
- Shaw DE,
- Ng YK,
- Moseley MJ
- Schalij-Delfos NE,
- Zijlmans LM,
- Wittebol-Post D,
- Tan KEWP,
- Cats BP
- Holmström G,
- el Azazi M, Jacobson L, Lennerstrand G
- Keith CG,
- Doyle LW
- ↵American Academy of Pediatrics and American College of Obstetricians and Gynecologists. Clinical considerations in the use of oxygen. In: Guidelines for Perinatal Care. 3rd ed. Elk Grove Village, IL: American Academy of Pediatrics; Washington, DC: American College of Obstetricians and Gynecologists; 1992:201–203
- Javitt J,
- Cas RD,
- Chiang Y
- ↵Armitage P, Berry G. Statistical Methods in Medical Research. 3rd ed. Oxford, England: Blackwell Science Inc; 1994:121
- ↵Efron B, Tibshirani RJ. An Introduction to the Bootstrap (Monographs on Statistics and Applied Probability, No. 57). New York, NY: Chapman & Hall; 1994
- Lucey JF,
- Dangman B
- Page JM,
- Schneeweiss S,
- Whyte HE,
- Harvey P
- Copyright © 1998 American Academy of Pediatrics