Significant correlations were found between mean spherical equivalents and stage of ROP. D indicates diopters.
The degree of astigmatism increased significantly with the increasing stage of ROP. D indicates diopters.
Holmström GE, Källen K, Hellström A, Jakobsson PG, Serenius F, Stjernqvist K, Tornqvist K. Ophthalmologic Outcome at 30 Months’ Corrected Age of a Prospective Swedish Cohort of Children Born Before 27 Weeks of GestationThe Extremely Preterm Infants in Sweden Study . JAMA Ophthalmol. 2014;132(2):182-189. doi:10.1001/jamaophthalmol.2013.5812
Copyright 2014 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
Follow-up at 30 months’ corrected age reveals eye and visual problems in one-third of children born extremely prematurely (<27 weeks’ gestation).
To investigate the ophthalmologic outcome of extremely preterm children at 30 months’ corrected age.
Design, Setting, and Participants
A prospective, population-based follow-up study (Extremely Preterm Infants in Sweden Study [EXPRESS]) was conducted in Sweden. The population included extremely preterm infants (<27 weeks’ gestation) born in Sweden between 2004 and 2007, of whom 491 survived until age 2.5 years. Screening for retinopathy of prematurity (ROP) was performed in the neonatal period. At 30 months’ corrected age, an ophthalmologic assessment was performed in 411 of 491 children (83.7%).
Main Outcomes and Measures
Visual acuity, manifest strabismus, and refractive errors were evaluated.
Visual impairment was identified in 3.1% of the children, and 1.0% were blind. Refractive errors, defined as myopia less than −3 diopters (D), hypermetropia greater than +3 D, astigmatism 2 D or more, and/or anisometropia 2 D or more, were found in 25.6% of the children, and 14.1% had manifest strabismus. There were significant associations between visual impairment and treated ROP (P = .02), cognitive disability (P < .001), and birth weight (P = .02). Multiple regression analyses revealed significant associations between strabismus and treated ROP (P < .001), cognitive disability (P < .01), and cerebral palsy (P = .02). Refractive errors were significantly correlated with severity of ROP (right eye, P < .001; left eye, P < .01). Children who had been treated for ROP had the highest frequency (69.0%) of eye and visual abnormalities.
Conclusions and Relevance
One-third of the extremely prematurely born children in this study had some kind of eye or visual problems, such as visual impairment, strabismus, or major refractive error. Despite being born extremely preterm, the present cohort has a similar prevalence of blindness and visual impairment as in previous Swedish cohorts of children born less prematurely.
Various neurologic, cognitive, and behavioral sequels are well-known complications of premature birth.1- 3 Ophthalmologic problems are also well known, with retinopathy of prematurity (ROP) being the major threat in the neonatal period and an important cause of childhood blindness worldwide.4 In older children, an increase in strabismus and refractive errors, as well as deficiencies in visual acuity, contrast sensitivity, stereopsis, visual fields, accommodation, and visual perception, have been shown in various long-term population-based studies.5- 10
Modern neonatal care is continuously improving, providing us with a new population of survivors born extremely preterm. The long-term outcome of this group of children is of utmost interest. A prospective national study (Extremely Preterm Infants in Sweden Study [EXPRESS])11- 13 of Swedish infants with a gestational age (GA) of less than 27 weeks born during 2004-2007 revealed high survival (70%) and high incidences of severe (35%) and treatment-requiring (20%) ROP. In an ongoing, long-term follow-up of this cohort, the neurologic and developmental outcome at 30 months’ corrected age has recently been reported.14 The present study aimed to investigate the ophthalmologic outcome of this group of extremely preterm children.
During a study period from April 1, 2004, to March 31, 2007, 707 infants with GA at birth less than 27 weeks (<27 + 0) were born alive; 506 survived to be included in studies of ROP in the neonatal period.12,13,15,16
Of the original 707 infants in the study cohort, 497 infants (70.3%) were alive at 1 year11 and, at 2.5 years’ corrected age, 6 of those children had died. Of the 491 remaining in the cohort, 411 children (83.7%) had an eye assessment performed by pediatric ophthalmologists and were included in the present study group. The study was approved by the ethics committee, medical faculty, of Lund University, Lund, Sweden. Oral informed consent for data acquisition was provided by the parents. No financial compensation was provided.
Ophthalmologic examination was scheduled at 2.5 years’ (±3 months) corrected age. The following variables were evaluated.
Best-corrected vision was assessed binocularly and monocularly, if possible. Three different tests with gradually decreasing difficulty were used: (1) ability to identify single optotypes 0.4 Lea Hyvärinen (LH) test at 3 m distance, (2) ability to fixate and follow a toy of 5 cm at 30 cm, and (3) ability to fixate and follow a light/torch at 30 cm. Children or eyes that were not able to identify an optotype at 3 m or a toy at 30 cm were considered to have impaired vision. Children or eyes that were not able to fixate and follow a light were considered to be blind.
Ocular motility and nystagmus were assessed and a cover test was performed for near (30 cm). Strabismus was defined as “manifest strabismus” (ie, esotropia, exotropia, or hypertropia/hypotropia). Evaluation of stereopsis was performed with the Lang test.
Refraction was assessed with retinoscopy in cycloplegia (cyclopentolate hydrochloride, 0.85%, and phenyl hydrochloride, 1.5%). Automatic refraction (Retinomax; Nikon) in cycloplegia was performed if retinoscopy was not possible.
Refraction was calculated as spherical equivalents and was evaluated as continuous values. For comparison with other studies, hypermetropia was regarded as significant if more than +3 diopters (D) and myopia was divided into categories of less than 0 D and less than −3 D (high myopia). Astigmatism was recorded as negative cylinders and was divided into categories of 1 D or more and 2 D or more, respectively. Anisometropia was divided into categories of 1 D or more and 2 D or more.
A composite score (eye and visual problems) was defined as visual impairment of the worse eye, strabismus, or refractive errors (ie, myopia, <−3 D; hypermetropia, >+3 D; astigmatism, ≥2 D; and/or anisometropia, ≥2D).
The exterior eye and the anterior segment were evaluated. Furthermore, ophthalmoscopy with fundus examination was performed.
Children with cerebral palsy (CP) were divided into 3 groups: mild (walking without an aid), moderate (walking with an aid), and severe (unable to walk even with an aid).14 Cognitive function was assessed with the Bayley Scales of Infant and Toddler Development, Third Edition (Bayley-III).17 A cognitive score was calculated and compared with the mean (SD) of a control group of children born at term.14 The results of the cognitive scores were divided into 4 groups: no disability, mild disability (<−1 SD to −2 SD below controls), moderate disability (<−2 SD to −3 SD below controls), and severe disability (<−3 SD below controls). Bayley-III testing had not been performed in 35 of the 411 children, but in 21 of these children information regarding cognitive function was extracted from patients’ medical records.
Possible associations between visual impairment (blindness or able to see only a light), strabismus, or eye and visual problems (composite variable, as described above) and possible risk factors (sex, GA, birth weight [BW], small for GA [SGA], standard deviation of BW (SDBW), SD from expected weight at 36 weeks’ GA [SD36wk], stage of ROP/treated ROP, any CP, and moderate or severe cognitive disability) were investigated using univariate and multiple logistic regression analyses. All variables with P < .20 were initially included in the multiple models and excluded if the P values in the multiple models exceeded .20. In these analyses, the number of independent variables never exceeded one-tenth of the number of cases. Therefore, when the numbers were small, only univariate analyses were performed. Statistical analyses were performed using Gauss, version 10 (Aptech Systems Inc, http://www.aptech.com).
In the study group of 411 children, 55.7% were boys (n = 229) and 44.3% were girls (n = 182). Mean gestational age was 25.4 weeks (range, 22.1-26.9 weeks), and mean BW was 783 g (range, 348-1315 g). Retinopathy of prematurity was found in the neonatal period in 73.7% of the infants (n = 303); 38.5% (n = 158) had mild ROP (stages 1 and 2), 35.2% children (n = 145) had severe ROP (stages 3-5), and 20.4% children (n = 84) had been treated for ROP. The distribution of ROP in relationship to GA at birth is reported in Table 1.
Eye assessment had been performed at a mean corrected age of 30.5 months (range, 20.4-45.9 months). In 90.8% of the children (364 of 401 with available information of time at assessment), examinations had been done within 2.5 years’ (± 3 months) corrected age. Major ophthalmic outcomes are presented in Table 1.
Vision was assessed binocularly in 94.9% of the children (n = 390) (Table 2). Of these, 51.0% of the children (n = 199) were able to identify a single optotype (0.4) binocularly at 3 m, and 45.9% of the children (n = 179) could fixate and follow a toy of 5 cm at 30 cm. Twelve children (3.1%) could neither identify an optotype nor fixate or follow a toy and were regarded as visually impaired. Of these 12 children, 8 could only fixate and follow a light and 4 were blind. Nine of the 12 visually impaired children, including the 4 who were blind, had been treated for ROP. Distribution of visual impairment in relation to GA at birth is reported in Table 1.
Monocular vision could be assessed in 83.5% of the children (n = 343) (Table 2). Analysis of the better eye of each child revealed that 5.8% (n = 20) could not identify an optotype or fixate and follow a toy. Eleven of these eyes, including the 4 blind eyes, had been treated for ROP.
Regarding the worse eye of each child, 7.0% (n = 24) could not see an optotype or a toy (Table 2). Thirteen of these 24 eyes had been treated for ROP.
Any CP was diagnosed in 6.6% (n = 27) of the 411 children, and 3.7% (n = 15) had moderate or severe CP. Moderate or severe cognitive disability (<−2 SD below controls) was diagnosed in 11.4% (43 of 376) of the tested children and in 14.3% of the 21 children (n = 3) whose information regarding cognitive function was extracted from medical records.
Quiz Ref IDLogistic regression analyses regarding associations between visual impairment (blindness or able to only fixate and follow a light) and sex, GA, BW, SGA, SDBW, SD36wk, stage of ROP/treated ROP, any CP, and moderate or severe cognitive disability were performed. There were significant associations between visual impairment and BW, severe treated ROP, and cognitive disability (Table 3). A multiple regression analysis was not possible because of the small number of children (n = 12) with visual impairment.
Nystagmus was found in 19 of the 399 children (4.8%) who had been assessed for that condition. Thirteen of these children had ROP of stage 3 or higher. Eleven of the children with nystagmus (57.9%) had moderate to severe CP or moderate to severe cognitive disability. Strabismus was found in 13 of the children, all of whom had an esotropia.
Sixteen children were registered at low-vision centers (Table 4). In 10 of the children, the cause of visual impairment was regarded as mainly retinal; in 4, cerebral; in 1, combined retinal/cerebral; and in 1, unknown.
Quiz Ref IDManifest strabismus was found in 14.1% of the 388 assessed children (n = 55). Eighty percent (n = 44) of those with strabismus had an esotropia and 20% an exotropia (n = 11). Forty-seven of the 55 children with strabismus (85.5%) had ROP in the neonatal period and 34 (61.8%) had stage ROP 3 or more, of whom 26 had received treatment for ROP. The frequency of strabismus was reduced with increasing GA at birth (Table 1).
Logistic regression analyses regarding associations between manifest strabismus and sex, GA, BW, SGA, SDBW, SD36wk, stage of ROP and treated ROP, any CP, and moderate or severe cognitive disability revealed significant associations with GA, BW, SD 36wk, severe treated ROP, CP, and cognitive disability (Table 3). In a multiple regression analysis, only severe treated ROP, CP, and cognitive disability were significantly correlated with strabismus (Table 3).
The Lang stereo test was performed in 316 children. A positive response was recorded in 79.4% of these children (n = 251), and stereopsis was either absent or could not be reliably evaluated in the remaining 65 children. The frequency of a positive Lang test was reduced with decreasing GA at birth, from 84% in children born with a GA of 26 weeks to 67% in those with a GA of 22 to 23 weeks.
Ocular motility was evaluated in 399 children, of whom 4.5% (n = 18) were regarded to have pathologic motility. All 18 of them had esotropia; 4 had overactions of the inferior obliques, and 10 had abduction deficiencies. In the remaining 4 children, the type of abnormality was not described.
Refraction in cycloplegia was assessed in 394 children (393 right and left eyes, respectively). The results reported below are based on manual retinoscopy (the criterion standard) in 303 children and on automatic retinoscopy (Retinomax) in the remaining 91 children.
Mean spherical equivalent of the right eyes was +1.45 D (range −9.5 to +8) and of the left eyes +1.42 D (range, −12 to + 8.5). The distribution of spherical equivalent of the right eyes in relationship to stage of ROP is presented in Figure 1. A hypermetropia of 5 D or more was found in 3.8% of the 393 right eyes (n = 15) and 4.1% of the left eyes (n = 16), and a myopia of 6 D or more was found in 2.5% (n =10) and 3.1% (n = 12) of the right and left eyes, respectively.
The mean spherical equivalents of both eyes were reduced, but not significantly, with decreasing GA at birth. There were significant correlations between mean spherical equivalents and stage of ROP (right eye, P < .001; left eye, P < .01), that is, the higher the stage of ROP, the lower the mean spherical equivalent (toward myopia).
Mean astigmatism of the right and left eyes was −0.58 D (0 to −6). The distribution of astigmatism of the right eyes in relationship to stage of ROP is illustrated in Figure 2. The degree of astigmatism increased with increasing stage of ROP (right eye, P < .01; left eye, P < .001).
Mean anisometropia was 0.54 D. In 6.4% of the 392 children (n = 25), there was an anisometropia of 1 D or more but less than 2 D, and 7.7% of the children (n = 30) had an anisometropia of 2 or greater. The degree of anisometropia increased with decreasing GA (P = .01) and with increasing stage of ROP (P < .001).
Refractive errors, defined as myopia less than −3 D, hypermetropia greater than +3 D, astigmatism 2 D or more, and/or anisometropia 2 D or more, were found in 25.6% (101 of 394) of the right or left eyes, which had been refracted in cycloplegia. Furthermore, refractive errors, as defined above, were found in 14.4% of eyes without ROP (15 of 104), 18.2% in those with mild ROP (28 of 154), 18.6% in those with severe untreated ROP (11 of 59), and 61.0% in those with treated ROP (47 of 77). The distribution of refractive errors in relationship to GA is reported in Table 1.
Ninety-two of the 411 children (22.4%) were wearing glasses. These had been prescribed previously or at the examination.
Bilateral anterior polar cataracts were seen in 1 child and monocular cataracts were identified in 2 children. One child had glaucoma of the right eye (blind in both eyes owing to stage 4B ROP). Six children had ptosis in 1 eye. Five children had a diagnosis of optic atrophy and 1 had optic hypoplasia in both eyes. Macular heterotopia was noted in 10 right eyes and 6 left eyes, and retinal folds were noted in both eyes in 3 children.
Eye and visual problems according to our definition were found in 33.3% of the total group of children (137 of 411) and in 20.4% in those without ROP, in 25.3% in those with mild ROP, in 27.9% in those with severe untreated ROP, and 69.0% in those with treated ROP. The relationship between these problems and GA is reported in Table 1.
Logistic regression analyses regarding associations between eye and visual problems and sex, GA, BW, SGA, SDBW, SD36wk, stage of ROP/treated ROP, any CP, and moderate or severe cognitive disability revealed significant associations with GA, BW, severe/treated ROP, CP, and cognitive disability (Table 3). In a multiple regression analysis, only severe treated ROP and cognitive disability were significantly correlated with eye and visual problems (Table 3).
Quiz Ref IDThis study at 2.5 years’ corrected age is of a new population of children, ie, children born extremely preterm. The findings revealed impaired vision (binocularly blind or fixates or follows a light) in 3.1%, strabismus in 14.1%, and refractive errors (ie, myopia <−3 D, hypermetropia >+3 D, astigmatism ≥2 D, and/or anisometropia ≥2 D) in 25.6% of the children. Together, 33.3% of the children had some kind of eye and visual problem, defined as visual impairment of the worse eye, strabismus, or refractive error. As expected, children who had been treated for ROP had by far the highest frequency (69.0%) of eye and visual abnormalities.
The study has the strength of being national and population-based. Furthermore, the infants were screened for ROP in the neonatal period,12 and the follow-up was organized and performed by pediatric ophthalmologists.
Only 1.0% of the study cohort was regarded as blind and 3.1% as visually impaired (ie, blind or only able to fixate or follow a light) when tested binocularly. Despite the early age, monocular testing could be performed in 83.5% of the children, of whom 2.3% were blind and 7% visually impaired in the worse eye. Population-based studies in Finland reported blindness in 1.5% of 211 children with BW less than 1000 g,19 in Great Britain in 2% of children less than 26 weeks’ GA,2 and in Norway in 2% of 306 children with less than 28 weeks’ GA.20 Hence, the prevalence of blindness in the Swedish population of children born at less than 27 weeks’ gestation seems to be similar to that in comparable studies.21 The prevalence of blindness and visual impairment in the present cohort is also similar to that of a previous population-based study22 of 260 children with a BW of 1500 g or less born in the Stockholm County of Sweden in 1988-1990. Because children in the present study were much more immature, one would have expected a higher prevalence of blindness.
The prevalence of manifest strabismus was 14.1% in the present study. Comparable population-based studies of extremely preterm children with similar ages and examined by ophthalmologists or orthoptists are difficult to find. The Stockholm study of children with a BW of 1500 g or less revealed a similar prevalence (13.5%) of strabismus at 2.5 years.22 The prevalence of strabismus in prematurely born children in the 2 Swedish cohorts cannot be compared with children born at term and of the same age, since such children were not included in those studies. The prevalence, however, is much higher than in the healthy Swedish population, with reported prevalences of 3.1% to 3.2% at 10 years.23,24 Treated ROP, CP, and cognitive disability were found to be the most important risk factors in the present study. We were not able to correlate strabismus with cerebral abnormalities, however, because magnetic resonance imaging was not routinely performed.
Refractive errors, defined as myopia less than −3 D, hypermetropia greater than +3 D, astigmatism 2 D or more, and/or anisometropia 2 D or more, were found in 25.6% of the study group. Guidelines for prescription of glasses had not been recommended before the 2.5-year examinations; consequently, conclusions regarding the frequency of glasses cannot be made. However, 22.4% of the children in the present study were wearing glasses or were prescribed glasses at the 2.5-year examination, which concords rather well with the above definition as well as the frequency of refractive errors (25.6%). The American Academy of Ophthalmology preferred practice pattern for prescription of glasses at similar ages uses a higher level of hypermetropia (>+4.5 D).25 Had their definition been used in the present study, it would have resulted in a lower prevalence of refractive errors (16.8%) and a lower rate of prescribed glasses.
The degree of refractive errors (ie, myopia, astigmatism, and anisometropia) was significantly correlated with increasing stage of ROP and treated ROP. Comparison with the previous Stockholm study of less immature children26 revealed similar to slightly lower prevalence of myopia (<−3 D) (3.1%), astigmatism (≥2 D) (4.8% right eyes/5.3% left eyes), and anisometropia (≥2 D) (4.9%) in that study. It could be speculated that the more central ROP in the most immature infants might affect the growth of the eye to a higher extent than in less central ROP, leading to higher prevalences of high myopia and astigmatism. This hypothesis, however, was not confirmed in the present study or in previous follow-up studies at age 3 years regarding myopia27 or astigmatism.28
As expected, treated ROP was a significant and important risk factor for visual impairment and strabismus in the present cohort as well as for refractive errors, in particular, high myopia, astigmatism, and anisometropia. This is illustrated by the fact that 69.0% of the children who had received treatment for ROP had some kind of eye and visual problem compared with 20.4% in those without ROP. Furthermore, CP was associated with strabismus and cognitive disability (evaluated with the Bayley test III) and was significantly correlated with visual impairment and strabismus, a finding that has been reported in previous studies.8- 10,29 Magnetic resonance imaging was not routinely performed; therefore, our findings cannot be correlated with possible cerebral abnormality. The well-known combination of retinal and cerebral causes for visual dysfunction in prematurely born children today30 is illustrated in the 16 study children registered at low-vision centers, 10 of whom had a mainly retinal source, 4 had a cerebral source, 1 probably had a combination of both, and the cause was unknown in the remaining child (Table 4). Furthermore, the correlations between cognitive defects and visual impairment and strabismus in the present study cohort emphasize the interaction between cerebral function and outcome of various visual functions.
Visual and ophthalmic evaluation at 2.5 years is not ideal but had to be coordinated with neurologic and psychological follow-up within the EXPRESS study.14 Surprisingly, approximately 95% of the cohort managed to take part in some of the ophthalmic evaluation, revealing that 1.0% were blind, 3% had visual impairment, and 33.3% had some type of eye or visual problem. Major visual dysfunction has probably already been detected at this age or before, which is in line with the opinion of Roberts et al,31 who investigated the stability of diagnosis of developmental disability from ages 2 to 8 years. However, the figures of the visual outcome at 2.5 years’ corrected age in the present study are probably only the “tip of the iceberg.” At this age, testing is rather crude, and more sophisticated tests to evaluate and detect more subtle visual perceptual dysfunctions affecting the quality of everyday life cannot be performed until later.
Quiz Ref IDIt has been questioned whether the increasing survival of the new population of extremely prematurely born infants takes place at the expense of increased neurodevelopmental impairments. The present study cohort including infants of less than 27 weeks’ GA at birth had a high survival rate of 70%,11 a very high incidence of ROP (74%), and a high frequency of ROP treatment (20%).12,13 Nevertheless, comparison with the previous population-based study22,26 of less prematurely born infants in Sweden revealed similar prevalences of blindness, strabismus, and refractive errors as in the present study. Regardless of the unique and extremely preterm cohort, only 1.0% of the children in the present study were blind and 3.1% were visually impaired. However, one-third of the children had some kind of eye or visual problems, defined as visual impairment of the worse eye, strabismus, or major refractive error. Furthermore, more subtle defects will probably be detected when the children become older. Ongoing evaluation at 6.5 years will provide a more complete picture of the total visual function and will be useful when predicting and discussing outcome and follow-up recommendations. This issue was not the purpose of the present study.
Submitted for Publication: February 22, 2013; final revision received June 20, 2013; accepted June 26, 2013.
Corresponding Author: Gerd E. Holmström, MD, PhD, Department of Neuroscience/Ophthalmology, Uppsala University, S-751 85 Uppsala, Sweden (email@example.com).
Published Online: December 5, 2013. doi:10.1001/jamaophthalmol.2013.5812.
Author Contributions: Drs Holmström and Källen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: All authors.
Acquisition of data: All authors.
Analysis and interpretation of data: All authors.
Drafting of the manuscript: Holmström.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Holmström, Källen.
Obtained funding: Holmström, Källen, Hellström, Serenius, Stjernqvist, Tornqvist.
Administrative, technical, and material support: All authors.
Study supervision: All authors.
Conflict of Interest Disclosures: None reported.
Funding/Support: The study was supported by the Birgit and Sven Håkan Olsson Foundation, the Evy and Gunnar Sandberg Foundation, the “Lilla Barnets Fond” Children’s Fund, the Nordströmer Foundation, a research grant from Region Skåne, Stiftelsen för synskadade i fd M-län, the Swedish Association of the Visually Impaired, and grants 2006 to 3855 from the Swedish Research Council.
Role of the Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: The following members of the Extremely Preterm Infants in Sweden Study (EXPRESS) Group contributed to the study: Mats Blennow, MD, PhD, and Mikael Norman, MD, PhD (Department of Clinical Science, Intervention and Technology, Karolinska Instututet, Stockholm), Uwe Ewald, MD, PhD, Gunnar Sjörs, MD, PhD, and Bo Strömberg, MD, PhD (Department of Women’s and Children’s Health, Uppsala University, Uppsala), Vineta Fellman, MD, PhD (Department of Pediatrics, Clinical Sciences Lund, Lund University, Lund), Eva Lindberg, MD, PhD (Department of Pediatrics, Örebro University, Örebro), Pia Lundqvist, RN, PhD (Department of Health Sciences, Lund University, Lund), Karel Maršál, MD, PhD (principal investigator) and Grozda Pajic (Department of Obstetrics and Gynecology, Clinical Sciences, Lund University, Lund), Elisabeth Olhager, MD, PhD (Department of Clinical and Experimental Medicine, Linköping University, Linköping), and Lennart Stigson, MD (Department of Pediatrics, Institute of Clinical Sciences, Sahlgrenska Academy, Gothenburg University, Gothenburg), Sweden. In addition, Gunnar Lindgärde, MD (St Eriks Eye Hospital, Stockholm) and Kent Johansson, MD (Norrland University Hospital, Umeå) provided valuable help with examinations and data input. The following pediatric ophthalmologic colleagues participated in the 2.5-year follow-up phase: Pia Agervi, MD, PhD, Bahra Arif, MD, Ewa Czechowska, MD, Thomas Krebser, MD, Stefan Löfgren, MD, PhD, Kristina Teär-Fahnehjelm, MD, PhD, and Agneta Wallin, MD (St Eriks Eye Hospital, Stockholm), Birgitta Carlsson, MD (University Hospital, Örebro), Ylva Friberg Riad, MD (Skaraborgs Sjukhus, Skövde), Lena Hilmertz, MD (Centralsjukhuset, Karlstad), Anna-Lena Hård, MD, PhD (The Queen Silvia Children’s Hospital, Sahlgrenska Academy, University of Gothenburg, Gothenburb), Eva Larsson, MD, PhD, and Hanna Åkerblom, MD (Department of Neuroscience/Ophthalmology, University Hospital, Uppsala), Birgitta Sunnqvist, MD (Ryhov Hospital, Jönköping), and Janina Waga, MD (University Hospital, Lund), Sweden. The contributors received no financial compensation.