PEDIATRICS Vol. 107 No. 4 April 2001, pp. 656-659

Screening Guidelines for Retinopathy of Prematurity: The Need for Revision in Extremely Low Birth Weight Infants

Muhammad Subhani, MD^*, Adriann Combs, RNC, Pamela Weber, MD^§, Corina Gerontis, MD^§, and Joseph D. DeCristofaro, MD^*

From the Departments of ^* Pediatrics,  Nursing, and ^§ Ophthalmology, SUNY at Stony Brook, Stony Brook, New York.

	ABSTRACT

Top Abstract Methods Results Discussion Conclusion References

Objective.  To determine whether significant retinopathy of prematurity (ROP) can be detected before 31 to 33 weeks' postmenstrual age (PMA) in extremely low birth weight (ELBW) infants.

Methods.  Medical records of all ELBW infants (<1000 g at birth) admitted to our regional perinatal center between April 1995 and January 1999 were reviewed retrospectively. Screening examinations for ROP were routinely performed at 4 to 6 weeks' chronological age (CA) from birth and followed at least every other week. Data were collected for infants who developed ROP. We determined the PMA at which the first screening eye examination demonstrated prethreshold disease and the subsequent examination that showed threshold disease (if it occurred). The percentages of infants who developed prethreshold ROP diagnosed at 31 weeks' PMA and of those who progressed to threshold ROP before 34 weeks' PMA were determined.

Results.  All 258 ELBW infants were screened for ROP. Seventy-eight infants (30%) were diagnosed with prethreshold ROP. Twenty-seven of these infants (35%) progressed to threshold ROP. Ten infants who progressed to threshold ROP were <34 weeks' PMA at the time of this diagnosis. Of these 10 infants, 3 were diagnosed at 31 weeks' PMA, 4 at 32 weeks', and 3 at 33 weeks'.

Conclusion.  More than 80% of ELBW infants who developed prethreshold disease in this cohort were 33 weeks' PMA. The joint statement screening option of independently using 31 to 33 weeks' PMA for the first eye examination would have led to a diagnosis of threshold ROP on first examination in as many as 13% (10/78) of our patients. Early identification of prethreshold ROP is important for providing timely intervention in this rapidly progressive disease. Therefore, ELBW infants should receive initial ROP screening using the CA guideline of 4 to 6 weeks rather than the 31- to 33-week postconceptional age guideline.  Key words:  extremely premature infants, prethreshold ROP, threshold ROP.

The survival of extremely premature infants has increased dramatically over the last 2 decades.^1-4 However, such births are associated with significant long-term morbidity, including poor visual outcome.⁵ The prognosis for maintaining functional vision remains poor in advanced cases of threshold retinopathy of prematurity (ROP), even with the application of currently available treatments^6,7 (eg, laser treatment, cryotherapy, scleral buckling, and vitrectomy). Therefore, research efforts have gradually shifted from detecting and treating threshold ROP to diagnosing prethreshold disease and testing therapeutic interventions.^8,9

The current guidelines for providing the first ophthalmologic screening examinations in low birth weight premature infants are based on the 1997 recommendations of the joint statement from the American Academy of Pediatrics (AAP), the American Association for Pediatric Ophthalmology and Strabismus, and the American Academy of Ophthalmology. They suggest that for infants with birth weights of 1500 g or with a gestational age (GA) of 28 weeks, the screening should be done at a postmenstrual age (PMA) of 31 to 33 weeks or the chronological age (CA) of 4 to 6 weeks.¹⁰

Data from 2 previous decades were used in formulating these joint statement guidelines, representing an era when survival was uncommon in extremely low birth weight infants ([ELBW] infants, defined by the AAP as <1000 g or 23 to 25 weeks' GA at birth). Based on these guidelines, the timing of the initial ophthalmologic screening examination may vary widely between institutions caring for ELBW infants. For example, using the PMA criterion of 31 to 33 weeks for an infant born at 23 weeks' GA, the first screening examination would be performed at 8 to 10 weeks' CA after birth. This same infant would be screened earlier, at 27 to 29 weeks' PMA, using the CA criterion of 4 to 6 weeks. This 4-week difference may be crucial for detecting prethreshold ROP and preventing threshold disease detected at the initial screening examination.

For the present study, we hypothesized that if the first ophthalmologic screening examination in ELBW infants were performed at 31 to 33 weeks' PMA, a few infants with threshold ROP, rather than immature retina or prethreshold disease, would be detected. To test this hypothesis, we determined the frequency of ROP in a cohort of ELBW infants admitted to our nursery and examined using the CA criterion. Our purpose was to determine the frequency of occurrence of prethreshold and threshold ROP in ELBW infants at 31 to 33 weeks' PMA.

	METHODS

Top Abstract Methods Results Discussion Conclusion References

Study Participants

The University Medical Center at Stony Brook is a regional high-risk perinatal and neonatal referral facility. This retrospective review comprised all ELBW premature infants (<1000 g at birth) admitted to our neonatal intensive care unit who received ophthalmologic screening examinations for ROP from April 1995 through January 1999. Data were collected from the logbook for eye examinations in the neonatal intensive care unit, the ophthalmologist's records, and patients' medical charts. The frequency of ELBW births and admission rates remained constant during the study period.

Ophthalmologic Examinations Schedule A registered nurse (A.C.) scheduled all eye examinations for detecting ROP in ELBW infants at risk in our unit. GA was assessed using prenatal estimates using the date of the last menstrual period, prenatal sonogram (if available), and initial Ballard examination.¹¹ The attending neonatologist assigned the final GA based on this information. The initial ophthalmologic screening examinations in this study population were performed by a pediatric ophthalmologist using the joint statement guideline of 4 to 6 weeks' CA after birth. Subsequent examinations were arranged every 2 weeks, or sooner if recommended by the ophthalmologist.

Classification of ROP The International Classification of Retinopathy of Prematurity¹² was used to define the zones of ROP (Table 1). Note was made of the zones of vascularization, number of total and contiguous clock hours of ROP, and presence or absence of plus disease. The ophthalmology team consisted of 2 experienced ophthalmologists (P.W., who also was a retinal surgeon), who shared similar management strategies. Together they performed ~250 examinations per year for ROP and provided all interventional therapies for these patients.

	RESULTS

Top Abstract Methods Results Discussion Conclusion References

Of the 258 ELBW infants who were screened for ROP during the study period, 78 (30%) were diagnosed with prethreshold ROP. Demographic data for these infants are shown in Table 2. The earliest diagnosis of prethreshold ROP was made at 5 weeks' CA in the most immature infant (23 weeks' GA), and 32% (25/78) were diagnosed at 6 weeks' CA. The earliest PMA of prethreshold disease was 28 weeks in a 23-week GA infant. The incidence of prethreshold disease in ELBW infants diagnosed at 33 weeks' PMA was 24% (63/258). More than 80% (63/78) of all infants with prethreshold disease were diagnosed at 33 weeks' PMA (Table 3). The median (range) PMA at the onset of prethreshold ROP was 32 (28-36) weeks, at a median CA of 7.6 (5.1-12.3) weeks.

In our most immature infants (23-25 weeks' GA), we found prethreshold disease in 56 (Table 4). This represents 72% of the infants with prethreshold ROP. The median birth weight of these infants was <700 g. The median age of prethreshold diagnosis was 32 weeks' PMA in this subgroup, with a range of 5 to 12 weeks' CA and a median of 8 weeks' CA. Thus of the 258 ELBW infants screened for ROP, only 1 was diagnosed with prethreshold disease at 5 weeks' CA.

Progression to threshold ROP occurred in 35% (27/78) of infants with prethreshold ROP (Tables 2-4). The median (range) PMA at the onset of threshold ROP was 34.7 (31-40.7) weeks at a median (range) CA of 9.7 (8-13.7) weeks. There was no difference in the PMA at presentation of prethreshold disease (Table 2) between those who progressed and those who regressed. The most immature ELBW infants were found to have highest rate of progression to threshold disease, 22 of 27 or >80% (Table 4).

Ten of the 27 infants (37%) who progressed to threshold ROP were 33 weeks' PMA at diagnosis; all 10 infants had zone 1 disease. In fact, of all infants with zone 1 threshold disease, 43% (10/23) presented at 33 weeks' PMA. Four of the 27 infants who progressed to threshold disease developed stage 4 or 5 ROP; 3 of these were 33 weeks' PMA at the time of this diagnosis. Alarmingly, we found threshold ROP in 3 infants at 31 weeks' PMA. Using the joint statement screening option of 31 to 33 weeks' PMA for the first eye examination in our ELBW infant population would have led to a diagnosis of threshold ROP on the first examination in as many as 13% (10/78) of infants.

	DISCUSSION

Top Abstract Methods Results Discussion Conclusion References

Using the PMA criterion of 31 to 33 weeks alone for first examination is inappropriate and inadequate for screening to detect early ROP in infants <1000 g birth weight. ELBW infants in our cohort were diagnosed with prethreshold ROP as early as 5 weeks of age (Table 4). Prethreshold disease was diagnosed as early as 28 weeks' PMA and threshold disease as early as 31 weeks' PMA. Even if initial screening had been routinely performed at 31 weeks' PMA in this cohort (rather than 4-6 weeks' CA, as was done), we would have found prethreshold disease in 30 infants and threshold disease in 3 infants at the time of their first screening examination. Thus, 40% of ELBW infants who develop prethreshold ROP can be diagnosed by 31 weeks' PMA. Alarmingly, by 33 weeks' PMA, 10 infants had already been diagnosed with threshold ROP that warranted intervention. All 10 patients had zone 1 disease with plus disease. Any delay in diagnosis of threshold disease in this group would have led to a worse visual outcome. A delay in diagnosis of threshold disease would prevent earlier intervention in patients with posterior disease in whom prethreshold treatment may be necessary.

Survival of ELBW infants has steadily increased over the past 2 decades.^1-4 However, serious lifelong disabling morbidities persist, and ROP remains a major contributor.⁵ The timing of ROP screening examinations and the diagnosis of prethreshold ROP have been previously reported.^13,14 Hutchinson et al¹³ showed that using PMA alone may fail to diagnose ROP before threshold disease in the lowest birth weight infants (<750 g). We extend these findings into the ELBW group in infants as small as 480 g at birth. Indeed, the joint statement was based on data from larger infants and a time when survival of ELBW infants was far less common.

The frequency of ROP occurring at our institution was similar to that reported in other studies.^13,14 We found that 30% of ELBW infants developed prethreshold ROP; 35% of these progressed to threshold disease. The median onset of prethreshold disease was 32.3 weeks (Table 2), earlier than reported by Palmer et al.¹⁴ This difference may result from our narrow focus on infants who are of lower GA and birth weight. Alternatively, significant changes in neonatal care have occurred since their study, almost 10 years ago. For example, antenatal steroids are now used with greater frequency and may have influenced our findings.

In our cohort, the PMA at the diagnosis of prethreshold ROP was 33 weeks in 81% of the infants. Perhaps more alarming, 10% of the infants who progressed to threshold ROP were 33 weeks' PMA at the time of diagnosis. Twenty-three of the infants who progressed to threshold disease presented with zone 1 disease. This is clearly different from previous reports and may be a unique characteristic of the ELBW population.

Additionally, ELBW infants may progress more rapidly from prethreshold to threshold disease than preterm infants, 1000 to 1500 g and 28 to 32 weeks' gestation at birth.¹³ The most immature infants in our cohort, 23 to 25 weeks' GA, had the greatest rate of progression, 80% (Table 4). Thus, the ideal time to screen for ROP in ELBW infants is when the onset of prethreshold ROP is detected. We speculate that earlier examination and diagnosis of prethreshold disease will allow new interventions and strategies to help prevent poor visual outcomes.

Prethreshold disease was diagnosed as early as 5 weeks' CA in ELBW infants but on average was detected by 8 weeks even in the most immature subgroup (23-25 weeks, Table 4). Therefore, if screening in ELBW infants were begun at 4 weeks' CA and repeated every 2 weeks, 50% of the infants with prethreshold ROP would be diagnosed by their third examination. On the other hand, if screening were begun at 6 weeks and repeated every 2 weeks, >75% would be diagnosed by the third eye examination. Thus, screening eye examinations starting from 4 weeks' CA would result in extra and probably unnecessary eye examinations. Our data suggest that screening eye examinations in ELBW infants should start at 5 to 6 weeks' CA or 42 days of life, not at the 31 to 33 weeks' PMA currently recommended. The joint guidelines must be clarified and updated in this new era of survival of the ELBW.

	CONCLUSION

Top Abstract Methods Results Discussion Conclusion References

The PMA criterion of 31 to 33 weeks for initial ROP screening should not be applied to the ELBW population. Rather, these infants should receive initial ROP screening using the CA criterion. In our cohort of ELBW infants, prethreshold ROP was present as early as 5 weeks' CA. We conclude that screening this group of infants should begin by 5 to 6 weeks (42 days) of age. We speculate that earlier examination and diagnosis of prethreshold disease will allow us to intervene with new treatments and strategies to decrease the rate of adverse visual outcome.

	ACKNOWLEDGMENT

We thank Dr Stephen Baumgart for his insightful review and critical analysis of this report.

	FOOTNOTES

Received for publication Mar 13, 2000; accepted Aug 7, 2000.

Reprint requests to (J.D.D.) Health Sciences Center T11-060, SUNY at Stony Brook, Stony Brook, New York, NY 11794-8111. E-mail: jdecrist{at}mail.som.sunysb.edu

	ABBREVIATIONS

ROP, retinopathy of prematurity; AAP, American Academy of Pediatrics; GA, gestational age; PMA, postmenstrual age; CA, chronological age; ELBW, extremely low birth weight.

	REFERENCES

Top Abstract Methods Results Discussion Conclusion References

1. O' Shea M, Klinpenter KL, Goldstein DJ, The changing pattern of neonatal mortality in a regionalized system of perinatal care: a current update. Pediatrics. 1999; 104:1064-1069 [Abstract/Free Full Text]
2. Herman T, Hein A, Lofgren MA Survival and developmental disability in infants with birth weights of 501 to 800 grams born between 1979 and 1994. Pediatrics. 1997; 100:982-986 [Abstract/Free Full Text]
3. Hack M, Freidman H, Fanaroff AA Outcomes of extremely low birth weight infants. Pediatrics. 1996; 98:931-937 [Abstract/Free Full Text]
4. La Pine TR, Jackson JC, Bennett FC Outcomes of infants weighing less than 800 grams at birth: 15 years experience. Pediatrics. 1996; 96:479-483 [Abstract/Free Full Text]
5. Hussain N, Clive J, Bhandari V. Current incidence of retinopathy of prematurity, 1989-1997. Pediatrics. 1999;104(3). URL: http://www.pediatrics.org/cgi/content/full/104/3/e26
6. Hirose T, Schepens CL, Katsumi O, Mehta MC. Open-sky vitrectomy for severe retinal detachment caused by advanced retinopathy of prematurity. In: Flynn JT, Tasman W, eds. Retinopathy of Prematurity. New York, NY: Springer-Verlag; 1992
7. Cryotherapy for Retinopathy of Prematurity Group The natural outcome of premature birth and retinopathy of prematurity: status at one year. Arch Ophthalmol 1994; 112:903-912 [Abstract/Free Full Text]
8. The STOP-ROP, Multicenter Study Group Supplemental therapeutic oxygen for prethreshold retinopathy of prematurity (STOP-ROP), a randomized, controlled trial. I: primary outcomes. Pediatrics. 2000; 105:295-310 [Abstract/Free Full Text]
9. Hardy RJ. Early Treatment for Retinopathy of Prematurity (ETROP) Manual of Procedures. Bethesda, MD: National Eye Institute; 2000. Grant 5U10EY012471-02
10. A joint statement of the American Academy of Pediatrics, the American Association for Pediatric Ophthalmology and Strabismus, and the American Academy of Ophthalmology Screening examination of premature infants for retinopathy of prematurity. Pediatrics. 1997; 100:273-274 [Free Full Text]
11. Ballard JL, Khoury JC, Wedig K, Wang L, Eilers-Walsman BL, Lipp R New Ballard score expanded to include extremely premature infants. J Pediatr 1991; 119:417-423 [CrossRef][Medline]
12. Committee for the Classification of Retinopathy of Prematurity An international classification of retinopathy of prematurity. Arch Ophthalmol 1984; 102:1130-1134 [Abstract/Free Full Text]
13. Hutchison AK, Saunders RA, Neil JW, Lovering A, Wilson ME Timing of initial screening examinations for retinopathy of prematurity. Arch Ophthalmol 1998; 116:608-612 [Abstract/Free Full Text]
14. Palmer EA, Flynn JT, Hardy RJ, Incidence and early course of retinopathy of prematurity. Ophthalmology. 1991; 98:1628-1640 [Medline]