Myopia of Prematurity

All content on Eyewiki is protected by copyright law and the Terms of Service. This content may not be reproduced, copied, or put into any artificial intelligence program, including large language and generative AI models, without permission from the Academy.

Article initiated by:

Michael Simmons, MD

All contributors:

S. Grace Prakalapakorn, MD, MPH, Michael Simmons, MD, Vandana Reddy, MD, Riccardo Vinciguerra, MD, Carla J. Osigian, MD, Bruno Battaglia, Brian Shafer, MD

Assigned editor:

Brian Shafer, MD, Carla J. Osigian, MD

Review:

Assigned status Up to Date

 by Brian.Shafer on August 7, 2026.

add
Contributing Editors:	add

Disease Entity

Myopia of prematurity (MOP) is a form of refractive error related to alterations in the development of the anterior segment, rather than axial length, that occur in individuals born prematurely. It is a distinct entity from pathologic or school-age myopia. MOP is closely linked to retinopathy of prematurity (ROP) and its treatment, although there is substantial evidence that premature individuals are at risk for myopic refractive error even in the absence of ROP.

Efforts to describe the role ROP has in MOP have led to a number of proposed names to distinguish whether myopia developed solely from prematurity (i.e., true "myopia of prematurity") or as a sequela of ROP with treatment ("myopia of retinopathy of prematurity") or without treatment ("myopia of spontaneously regressed ROP"). These terms are not universally accepted, however, and many publications do not differentiate between them.

MOP is clinically important because affected children are at risk of developing high myopia with an increased lifelong cumulative risk of myopia-related complications, amblyopia, strabismus, and reduced visual acuity.

Disease Classification

There is no specific ICD code for myopia of prematurity. Depending on the context, codes for retinopathy of prematurity (ICD-9: 362.20; ICD-10: H35.109), progressive high myopia (ICD-9: 360.21), degenerative myopia (ICD-10: H44.20), or simply myopia (ICD-9: 367.1; ICD-10: H52.13) may be appropriate. Likewise, there is no specific MeSH identifier for MOP.

Under the IMI framework, MOP falls within the category of secondary myopia in infants and young children—a group etiologically distinct from school-age myopia, in which prematurity and genetic/syndromic causes predominate.

History

Early studies on infants with ROP led to the observation that premature infants with and without ROP were predisposed to myopic refractive error, but the prevalence of myopia was much higher in infants with ROP. In 1981, Fledelius observed that "myopia of prematurity is almost obligatory in cases of incomplete cicatricial retrolental fibroplasia." (The term "retrolental fibroplasia" is a historic name for stage 5 ROP.)

The Cryotherapy for Retinopathy of Prematurity (CRYO-ROP) trial (study start 1986) included a natural history subgroup which helped conclusively demonstrate the reality of MOP. This and subsequent trials—including the Early Treatment for Retinopathy of Prematurity (ETROP) trial (2001) and the Bevacizumab Eliminates the Angiogenic Threat of Retinopathy of Prematurity (BEAT-ROP) trial (2008)—provided key information about MOP and the effects of ROP treatment (cryotherapy, laser photocoagulation, and intravitreal bevacizumab) on its development. More recently, the RAINBOW (ranibizumab) and FIREFLEYE/FIREFLEYE Next (aflibercept) programs have extended refractive outcome data to additional anti-VEGF agents.

Epidemiology

Landmark trial data

In the natural history population of the CRYO-ROP trial there was an overall prevalence of myopia (< −0.25 D) of 21% among all subjects at 1 year, but the prevalence of myopia at the same time point among the subgroup with severe ROP was 80%. Likewise, the overall prevalence of high myopia (< −5.0 D) was 3.9% at one year among all participants but was approximately 43% among those with severe ROP. Each 100 g decrease in birth weight correlated with a 10% increase in the prevalence of myopia.

In the ETROP study, which enrolled participants with prethreshold ROP, the prevalence of myopia and high myopia was approximately 65% and 35% respectively at 4 to 6 years of age.

CRYO-ROP natural history data at 5½ years demonstrate a clear dose–response relationship between acute-phase ROP severity and both the magnitude and the variability of subsequent refractive error, as shown in the following figure.

Figure 1.

Multicenter Trial of Cryotherapy for Retinopathy of Prematurity: Natural History ROP: Ocular Outcome at 5½ Years in Premature Infants With Birth Weights Less Than 1251 g. Arch Ophthalmol. April 30, 2002. Content used under license from the JAMA Network® © American Medical Association

Gradient of risk by ROP severity

The Indian Twin Cities ROP Study (ITCROPS, 838 preterm infants) quantified this gradient. Myopia (SER < −0.50 D) was present in 39.7% of infants with ROP versus 19.8% of infants without ROP, and prevalence increased with ROP stage: 19.7% (stage 1), 33.8% (stage 2), 45.6% (stage 3), and 59.5% (aggressive posterior ROP). Myopia prevalence doubled in treated versus untreated infants (54.5% vs 25.5%). The mean 1-year change in SER was −4.55 ± 1.38 D in APROP and −2.28 ± 0.57 D in stage 3 ROP, significantly greater than in other stages or in infants without ROP.

The EXPRESS study of extremely preterm children at 6.5 years similarly demonstrated progressive worsening of both spherical equivalent and visual acuity with increasing ROP severity and treatment status.

Figure 2. Distribution of Refraction, Expressed as Spherical Equivalent (Left) and Visual Acuity (Right), in Better Eyes in the 6-Year Extremely Preterm (EPT) Group and the Control Group Relative to Stage and Treatment of Retinopathy of Prematurity (ROP)

Ophthalmologic Outcome of Extremely Preterm Infants at 6.5 Years of Age. JAMA Ophthalmol. April 30, 2016. Content used under license from the JAMA Network® © American Medical Association

Mild and regressed ROP in more mature preterm infants

A 2026 retrospective cohort of 231 infants found that mid-preterm infants with a history of ROP—even mild or regressed disease—had a significantly more myopic refractive profile than mid-preterm infants without ROP (sphere P = .028; SE P = .012), whereas no significant differences were seen among late-preterm infants. No cases of high myopia occurred in this cohort. Mid-preterm infants with regressed ROP may represent an underrecognized group warranting targeted refractive surveillance.

Prematurity without ROP

Preterm infants without ROP also demonstrate abnormal refractive development. In a Nepalese cohort, spherical equivalent shifted from +0.84 ± 1.72 D at birth to −0.33 ± 1.95 D by 6 months, with significantly greater astigmatism and anisometropia than full-term infants (P < .01). In a longitudinal Scottish cohort of 59 preterm infants without ROP, myopia and anisometropia were associated with prematurity (P < .05), and 19% had clinically significant refractive errors at 4 years of age; notably, neither clinical risk factors nor early refractive measurements predicted refractive outcome at 4 years.

A 2025 PRISMA-based review of 32 studies (4,548 individuals) of preterm children without ROP found hyperopia to be highly prevalent (20.9%–86.2%), astigmatism in 14.1%–65.1%, and anisometropia in 1.4%–16.7%, with astigmatism predominating at age 1 year and hyperopia becoming the most common refractive error by age 6. Myopia rates remained relatively stable with age in this group. Anterior segment findings in infancy included greater central corneal thickness, smaller corneal diameter, reduced axial length, shallower anterior chamber depth, and increased lens thickness.

The following figure from a population-based Swedish study of 10-year-olds illustrates that although cryotreated children bear the greatest burden of myopia across all severity ranges, preterm children without ROP also develop myopia more often than term controls.

Figure 2.

A Population-Based Study of the Refractive Outcome in 10-Year-Old Preterm and Full-Term Children. Arch Ophthalmol. September 30, 2003. Content used under license from the JAMA Network® © American Medical Association

Variation by treatment modality

MOP prevalence and severity vary substantially with treatment modality. Laser photocoagulation results in a lower incidence of MOP than cryotherapy ablation.

The BEAT-ROP study compared outcomes following laser photocoagulation or intravitreal bevacizumab (IVB) and demonstrated a lower prevalence of myopia and lower spherical equivalents in IVB-treated patients. At age 2.5 years the mean spherical equivalent was −1.51 D in the IVB group versus −8.44 D in the laser group (P < .001) for zone I ROP. For zone II ROP, mean spherical equivalent was −0.58 D (IVB) versus −5.83 D (laser) (P < .001). Very high myopia (≥ −8.00 D) occurred in 3.8% of zone I and 1.7% of zone II eyes treated with IVB, in contrast to 51.4% and 36.4% of laser-treated eyes (P < .001). The study also found a positive correlation between the degree of myopia and the number of laser applications: −0.14 D per 100 laser shots.

The Cochrane review of anti-VEGF for ROP graded the BEAT-ROP refractive findings as low-quality evidence owing to a unit-of-analysis error (eyes rather than infants as the denominator) and risk of detection bias, an important caveat when counseling families. In that analysis, the risk of very high myopia (≥ −8 D) at 30 months was significantly lower with IVB (RR 0.06, 95% CI 0.02–0.20), and mean spherical equivalent was 5.68 D less myopic (95% CI 4.33–7.02).

Contemporary meta-analytic data

Study (year) Design / population Key refractive findings References Huang et al. (2026) 86 studies, 10,269 eyes; anti-VEGF vs laser vs cryotherapy vs vitrectomy Pooled mean SE: anti-VEGF −1.9 D, laser −3.8 D, cryotherapy −5.8 D, vitrectomy −6.3 D. High myopia (SE ≤ −5.0 D): 21.3% anti-VEGF vs 42.6% laser vs 55.4–58.6% vitrectomy/cryotherapy. Anti-VEGF vs laser RR for high myopia 0.39 (95% CI 0.25–0.61) [4] Kong et al. (2021) 13 studies, 1,850 eyes Anti-VEGF less myopic than laser (MD 1.80 D, 95% CI 0.97–2.63, P < .0001, I² = 78%); no significant difference in axial length, anterior chamber depth, or lens thickness [5] Tan et al. (2019) 2 RCTs + 5 nonrandomized studies; 272 IVB eyes, 247 laser eyes IVB associated with less myopic SEQ (P < .001), lower prevalence of high myopia (P < .05), and less astigmatism (P = .02) [6] Wang & Zhang (via AAP) 17 studies (13 nonrandomized); laser vs bevacizumab vs ranibizumab Similar efficacy and retreatment rates, but higher incidence of complications and myopia with laser (OR 0.38, 95% CI 0.19–0.75, P = .005) [7] [8-11]

The FIREFLEYE Next trial extended these observations to aflibercept, with high and very high myopia substantially more common after laser than after aflibercept at 2 years of chronological age, as illustrated below.

Figure 2. Cycloplegic Refraction at 2 Years of Chronological Age

Intravitreal Aflibercept vs Laser Therapy for Retinopathy of Prematurity: Two-Year Efficacy and Safety Outcomes in the Nonrandomized Controlled Trial FIREFLEYE next. JAMA Netw Open. March 31, 2024. Content used under license from the JAMA Network® © American Medical Association

Temporal pattern of development

Longitudinal studies indicate that the degree of myopia is not static from birth but develops over time, with the most rapid changes occurring during the first year of life.

In a prospective observational study of 64 preterm children followed from 0 to 7 years, refractive change followed a bilinear pattern in infants who required panretinal photocoagulation: myopic shift of −4.7 D/year before age 1.3 years, slowing to −0.15 D/year thereafter. In contrast, children with mild or spontaneously regressed ROP followed an essentially flat linear pattern (−0.004 D/year). Anisometropia in the severe ROP group increased approximately three times faster than in the mild/no-ROP group, and with-the-rule astigmatism increased significantly with age.

Progression into adolescence and adulthood

In a 17-year longitudinal study of laser-treated threshold ROP, all eyes assessed at age 17 years had myopia (mean SE −6.35 D; range −1.25 to −12.38 D), and 43% were highly myopic (SE < −6.0 D). These eyes had significantly poorer visual acuity (P < .001), greater cylinder (P < .001), higher corneal astigmatism (P < .001), flatter horizontal corneal curvature (P = .01), shallower anterior chamber depth (P < .001), thicker lens (P < .001), and shorter axial length (P = .021) than age-matched full-term controls. The authors concluded that myopia and astigmatism continue to progress through adolescence after school age and recommended long-term follow-up for both refractive status and anterior segment structural change.

A 2025 review of adult outcomes found that adults born preterm have a higher prevalence of refractive error, lower visual acuity, higher prevalence of strabismus, shorter axial length, and steeper corneal radius, with additional findings of increased macular thickness, thinner peripapillary RNFL, and altered foveal avascular zone anatomy. Adults with a history of ROP have a high risk of myopic refractive error, amblyopia, shallower anterior chambers, thicker crystalline lenses, higher corneal aberrations, and foveal hypoplasia.

Pathophysiology

The etiology of MOP is multifactorial, involving changes in corneal curvature, lens characteristics, and globe elongation. The defining pathophysiologic feature of MOP (with and without associated ROP) is abnormal development of the anterior segment. Eyes with MOP exhibit increased corneal curvature, thick lenses, and shallow anterior chambers. In contrast to pathologic myopia, where increased axial length is the hallmark, eyes with MOP characteristically have shorter axial lengths relative to their dioptric value.

Biometric evidence for the anterior segment mechanism

A cross-sectional A-scan study directly compared structural contributors to myopia in preterm versus full-term children (93 children, ages 2–13, all with ≥ −3 D myopia in at least one eye). Myopic preterm eyes had significantly steeper average keratometry than non-myopic full-term eyes (45.4 ± 0.4 D vs 43.5 ± 0.7 D, P = .008), shallower ACD than myopic full-term eyes (2.5 ± 0.5 vs 3.2 ± 0.3 mm, P = .01), thicker lenses (4.9 ± 1.0 vs 4.1 ± 0.3 mm, P = .001), and shorter axial length (P = .01) despite equivalent refractive error. The authors concluded that increased axial length drives myopia in full-term children, whereas corneal curvature and lens thickness are the major contributors in preterm children.

A cohort of 108 school-age children born at <35 weeks or <1500 g confirmed higher prevalences of myopia (48% vs 29%), hyperopia (23% vs 15%), and astigmatism (73% vs 41%) compared with age-matched national survey controls. Common ocular features were shallow ACD, thick lenses, and steep corneal curvature; hyperopic cases had the shortest axial length, whereas myopic cases had significantly shallower ACD and greater lens thickness. Children with a history of ROP had the most prominent anterior segment changes.

Notably, the Kong et al. meta-analysis found no significant difference in axial length, anterior chamber depth, or lens thickness between anti-VEGF– and laser-treated eyes despite a 1.80 D difference in spherical equivalent, indicating that the biometric mediators of the treatment effect remain incompletely characterized.

Proposed mechanisms

The exact mechanism whereby prematurity and ROP lead to the characteristic anterior segment aberrations of MOP is not known. Proposed explanations include:

Arrest of normal anterior segment development — Observations about the timing of MOP development and the improved refractive outcomes of laser over cryotherapy and IVB over laser have led some to suggest that MOP reflects a mechanical restriction of ocular growth.

Disruption of peripheral hyperopic defocus signaling — A recent cross-sectional study comparing laser-treated ROP eyes, untreated ROP eyes, preterm eyes without ROP, and full-term controls found that laser-treated eyes had significantly steeper keratometry, more myopic SE, shorter axial length, and reduced ACD compared with full-term controls. The authors hypothesize that laser therapy disrupts the peripheral hyperopic defocus mechanism, arresting axial elongation without a corresponding reduction in corneal and lenticular refractive power. The untreated (regressed) ROP group also showed significantly steeper keratometry and shorter axial length relative to controls, suggesting that this mechanism may be impaired both in infants with an immature peripheral retina and in those treated with laser.

Preservation of local growth factor signaling with anti-VEGF — Improved retinal vascular development with anti-VEGF treatment may allow more normal levels of local growth factors necessary for appropriate signaling cascades involved in anterior segment development, potentially accounting for the reduced occurrence of MOP in IVB-treated eyes.

Postnatal bone mineral deficiency

Retinal dysfunction — Abnormal retinal signaling in eyes with even mild ROP

Temperature interactions during ocular development

Risk Factors

Lower gestational age — Each decrement in gestational age is associated with more myopic SE and greater astigmatism.

Lower birth weight — In CRYO-ROP, each 100 g decrease in birth weight correlated with a 10% increase in myopia prevalence. Birth weight is inversely related to corneal astigmatism, corneal refraction, and corneal curvature.

Greater ROP severity — Stepwise increase in myopia prevalence from stage 1 through APROP.

Zone I disease — Independently associated with myopia and high myopia across treatment groups.

Treatment for ROP — The strongest single predictor. In the GPSY cohort, spherical equivalent was associated only with ROP treatment (β = −2.91, P < .001).

Treatment modality — Cryotherapy and vitrectomy > laser > anti-VEGF in myopic burden.

Number of laser applications — −0.14 D per 100 laser shots.

Diagnosis and Clinical Evaluation

Cycloplegic refraction is the cornerstone of diagnosis and should be performed serially. Handheld autorefraction confirmed by retinoscopy is commonly used in infancy.

Ocular biometry (axial length and keratometry) is important to distinguish refractive myopia associated with abnormal anterior segment development—the pattern typical of MOP—from axial myopia. The IMI guidance emphasizes that biometric evaluation is a priority in any infant or young child with high myopia, and that the presence of high myopia in this age group should prompt a search for an associated medical or syndromic diagnosis, potentially including ocular imaging, electrophysiologic testing, genetic testing, and involvement of pediatricians and clinical geneticists.

Associated findings that should be sought include astigmatism (often with-the-rule and progressive), anisometropia, amblyopia, strabismus (esotropia most common), nystagmus, macular dragging, and foveal hypoplasia. In children with regressed ROP, macular dragging (present in 33.6% of eyes in one series), smaller gestational age and birth weight, larger absolute spherical equivalent, myopia, astigmatism, and anisometropia were all significantly associated with poorer best-corrected visual acuity.

Differential Diagnosis

School-age (juvenile-onset) myopia — axial, later onset, normal anterior segment biometry

Pathologic myopia — marked axial elongation with myopic maculopathy

Syndromic/genetic high myopia — Stickler syndrome, Marfan syndrome, Weill-Marchesani syndrome, congenital stationary night blindness, retinal dystrophies

Lenticular myopia — anterior lenticonus, spherophakia, ectopia lentis, congenital cataract

Secondary myopia — medication-induced, posterior segment pathology

Management

Surveillance

Premature infants—and especially those with severe ROP or a history of ROP treatment—should be monitored vigilantly with serial cycloplegic refraction. Because the most rapid myopic shift occurs before approximately 1.3 years of corrected age in laser-treated infants, refraction should begin early and be repeated frequently during infancy.

Infants treated for ROP require lifelong eye examinations, particularly in the first 5 years of life. Reactivation of ROP is more common after anti-VEGF therapy, so meticulous arrangements for ongoing examinations after anti-VEGF treatment are critical to identify and manage cicatricial ROP and possible retinal detachment.

Given that myopia and astigmatism in laser-treated ROP eyes continue to progress through adolescence, follow-up should extend beyond school age and include monitoring of anterior segment structural change. Surveillance should also extend to preterm children without ROP, who have high rates of refractive error, anisometropia, and strabismus.

Optical Correction

Optical correction is essential and should be prescribed early, as infants treated with laser photocoagulation for severe ROP progress rapidly toward myopia and develop anisometropia and astigmatism with age. Correction may be more challenging and complex than in older children owing to high dioptric powers, anisometropia, associated nystagmus, and neurodevelopmental comorbidities.

Monitoring for amblyopia secondary to high myopia, anisometropia, or strabismus is essential. In the GPSY cohort, ROP treatment was associated with a markedly increased odds of amblyopia (OR 15.71, P ≤ .001), alongside gestational age ≤28 weeks (OR 2.92, P = .03) and placental insufficiency (OR 3.84, P = .01). Amblyopia therapy follows standard principles—refractive correction first, with patching or pharmacologic penalization as indicated.

Myopia Control

The IMI consensus notes that application of myopia control interventions in infants and young children with high myopia requires a case-by-case approach because of the lack of efficacy evidence in this specific age group; nearly all myopia control trials have enrolled school-age children with juvenile-onset axial myopia, a mechanistically distinct condition from MOP.

Key considerations when extrapolating from the school-age myopia control literature:

Mechanistic mismatch: Myopia control interventions target axial elongation. Because MOP is predominantly refractive (corneal and lenticular) rather than axial—with affected eyes typically having shorter-than-normal axial length—the theoretical benefit of axial-length-directed therapy is uncertain.

Evidence base in school-age myopia: The Cochrane living systematic review and network meta-analysis (104 RCTs, 17,509 children) found topical antimuscarinics and orthokeratology to be the most effective interventions. Ortho-K reduced axial elongation by −0.18 mm at 1 year and −0.30 mm at 2 years versus single-vision correction. Higher-dose atropine (≥0.1%) slows axial elongation by approximately 0.5 mm over 2 years but with more adverse events and possible rebound on cessation; the effect of low-dose atropine may be small and remains uncertain.

Practical barriers in this population: Ortho-K and soft multifocal contact lenses are frequently poorly tolerated (more than half of children discontinued in some trials) and may be impractical in infants and in children with neurodevelopmental comorbidity or nystagmus.

Preliminary data specific to MOP exist for spectacle-based interventions. A study of highly aspherical lenslet target (HALT) spectacle lenses reported reduced myopia progression in children with myopia of prematurity, though the evidence base remains small and requires confirmation in larger controlled studies.

Any decision to use myopia control therapy in a child with MOP should be individualized, made in conjunction with pediatric ophthalmology, and should not displace the primary priorities of full refractive correction and amblyopia treatment.

Refractive Surgery in Selected Cases

Keratorefractive surgery is an off-label use of an FDA-approved device in children and is reserved for those with amblyogenic anisometropia who fail or cannot tolerate conventional therapy—a scenario that may arise in MOP given the high rates of high myopia and anisometropia.

The AAO Ophthalmic Technology Assessment on laser refractive surgery for anisometropic amblyogenic refractive error in children summarizes the available evidence:

In a comparative case series of 27 children (ages 4–7) with myopic anisometropia >3 D and contact lens intolerance, PRK/LASEK reduced mean SE from −8.25 ± 2.37 D to −1.48 ± 1.13 D at 2 years, with mean decimal BCVA improving from 0.23 to 0.78 (versus 0.16 to 0.42 in a contact-lens-plus-patching control group). Notably, the proportion within ±1.00 D of target declined from 74% at 3 months to 53% at 2 years, reflecting significant regression.

In 18 children (mean age 6.7 years) with ≥4 D myopic anisometropic amblyopia who failed 6 months of occlusion, LASIK improved mean SE from −9.25 ± 3.43 D to −1.50 ± 1.23 D at 2 years and decimal BCVA from 0.26 to 0.82; however, eyes at target fell from 77.8% at 6 months to 55.6% at 2 years, with mean regression of −2.25 ± 1.7 D.

In 33 amblyopic patients (mean age 9.0 years) with ≥5 D myopic anisometropia, femto-LASIK/SMILE improved mean SE from −10.00 ± 2.39 D to −0.60 ± 1.43 D and logMAR BCVA from 0.98 to 0.41, with no lines of vision lost.

A 10-year retrospective series of 32 children (mean age 8.6 years) undergoing unilateral LASIK for myopic anisometropia >6 D after mandatory 6-month occlusion therapy reported mean SE improving from −10.3 ± 2.0 D to −1.3 ± 0.8 D, anisometropia decreasing from −9.5 ± 1.7 D to 1.0 ± 2.5 D at 10 years, BCVA improving from 0.04 to 0.6 decimal, restoration of orthophoria in all 11 patients with preoperative intermittent exotropia, and no post-LASIK ectasia on topography.

Important caveats specific to MOP: Refractive surgery in this population carries a risk profile greater than conventional therapy, requires general anesthesia in young children, is subject to substantial myopic regression during continued ocular growth, and must be considered against the abnormal corneal curvature and biometry characteristic of MOP eyes. Phakic intraocular lenses and refractive lens exchange are alternative approaches with documented success in improving acuity in children unable to wear spectacles.

Associated Conditions Requiring Concurrent Management

Strabismus — Approximately 80% of children with a history of severe ROP develop strabismus during the first 6 years of life.

Astigmatism and anisometropia — Both increase with age in severe ROP, with anisometropia progressing approximately three times faster than in mild/no ROP.

Cataract/lens opacification — Significantly greater in children treated for ROP.

Macular dragging and foveal hypoplasia — Present in a third of eyes with regressed ROP and strongly correlated with poorer BCVA.

Neurodevelopmental comorbidity — At age 8 years, a third of children with threshold ROP from CRYO-ROP required special education and nearly half had below-grade-level academic performance, which may limit compliance with optical correction and amblyopia therapy.

Prognosis

Visual prognosis depends heavily on ROP severity, treatment modality, and the presence of macular pathology rather than on refractive error alone.

In the ETROP study at 6 years, poor visual acuity was equally likely in treated and untreated eyes overall, though earlier treatment benefited type 1 disease (16% vs 25% unfavorable acuity; P = .004) but not type 2 disease.

At age 10 years in CRYO-ROP, approximately 25% of eyes in both treated and untreated groups had acuity of 20/40 or better, but fewer treated eyes were blind (33% vs 50%).

In a multicenter Japanese cohort of infants <500 g birth weight treated for type 1 ROP and followed 5 years, median BCVA was 0.15 logMAR, with 73% achieving ≥20/40 and 27% achieving ≥20/20. Median SE was −2.37 D; 75% had myopia and 25% high myopia. Anti-VEGF–treated eyes had less myopia (P < .009) with no difference in BCVA or astigmatism.

The following figure from the ETROP study shows the distribution of visual acuity categories at 6 years for early-treated versus conventionally managed eyes, illustrating that treatment shifts the overall distribution toward better function rather than simply preventing the worst outcomes.

Figure 4.

Final Visual Acuity Results in the Early Treatment for Retinopathy of Prematurity Study. Arch Ophthalmol. May 31, 2010. Content used under license from the JAMA Network® © American Medical Association

Counseling Points for Families

Myopia is expected after treated severe ROP and typically develops most rapidly in the first 12–18 months of corrected age.

Anti-VEGF treatment is associated with substantially less myopia than laser, though this evidence is graded as low quality by Cochrane, and anti-VEGF carries a higher rate of ROP reactivation requiring prolonged, reliable follow-up.

Glasses correct the refractive error but do not reverse the underlying anterior segment changes; ongoing amblyopia surveillance is required.

Refractive error and astigmatism continue to progress into adolescence, so follow-up must extend well beyond the preschool years.

Preterm children without ROP also warrant refractive screening, as they carry elevated rates of hyperopia, astigmatism, and anisometropia.

Additional Resources

American Academy of Ophthalmology. Amblyopia Preferred Practice Pattern.

American Academy of Ophthalmology. Retinopathy of Prematurity. EyeSmart/Eye Health.

American Academy of Pediatrics. Screening Examination of Premature Infants for Retinopathy of Prematurity (joint policy statement with AAO and AAPOS).