A Revision of the International Classification of Retinopathy of Prematurity

As in earlier versions of the classification, ICROP3 will advance clinical data collection, prompt investigation of the impact of these additional factors on the course of the disease, and improve communication between physicians.Retinopathy of prematurity (ROP) remains a serious risk to the vision of premature infants, continuing to cause permanent vision loss for some. A clinical classification system for acute ROP, the International Classification of Retinopathy of Prematurity (ICROP), was published in 1984 with input from experts from around the world.1Committee for the Classification of Retinopathy of PrematurityAn international classification of retinopathy of prematurity.Arch Ophthalmol. 1984; 102: 1130-1134Crossref PubMed Scopus (1121) Google Scholar ICROP created a semiquantitative description of the extent and severity of acute disease by documenting zone, stage, and the presence or absence of plus disease (posterior pole vascular dilation and tortuosity) in the retina immediately around the optic disc. The value of such a system to describe ROP and its severity was quickly demonstrated by its successful use in a multicenter clinical trial of cryotherapy for the prevention of retinal detachment and visual loss from ROP.2Cryotherapy for Retinopathy of Prematurity Cooperative GroupMulticenter trial of cyrotherapy for retinopathy of prematurity.Arch Ophthalmol. 1990; 108: 1408-1416Crossref PubMed Scopus (312) Google Scholar ICROP has also improved clinical care by allowing accurate documentation of ROP, promoting comparison between serial examinations even by multiple clinicians, and facilitating consultation with colleagues ever since. As in earlier versions of the classification, ICROP3 will advance clinical data collection, prompt investigation of the impact of these additional factors on the course of the disease, and improve communication between physicians. Fine-tuning of the ICROP has been undertaken in the past, most recently in 2005.3International Committee for the Classification of Retinopathy of PrematurityThe International Classification of Retinopathy of Prematurity revisited.Arch Ophthalmol. 2005; 123: 991-999Crossref PubMed Scopus (2057) Google Scholar The ICROP3 published this year updates the classification to better describe the disease as it is currently being managed.4Chiang M.F. Quinn G.E. Fielder A.R. et al.International Classification of Retinopathy of Prematurity, 3rd edition.Ophthalmology. 2021; 128: e51-e68Abstract Full Text Full Text PDF PubMed Scopus (60) Google Scholar The key updates in ICROP3 address concepts added over the last 2 decades that are thought to influence the outcome or are associated with changes specific to injection of anti-intravitreal vascular endothelial growth factor (VEGF). While not as revolutionary as the concept of zone, stage, and plus/not retinal vascularity in ICROP, these additional observations may materially affect the baby’s risk for an adverse outcome. As in earlier versions of the classification, ICROP3 will advance clinical data collection, prompt investigation of the impact of these additional factors on the course of the disease, and improve communication between physicians. The important changes are descriptions of a distinct posterior portion of zone II, the temporal notch, subcategories of stage 5, a continuous spectrum of vascular changes, aggressive ROP, and nomenclature describing ROP regression and reactivation of disease observed after anti-VEGF therapy. In my view, the 2 most important updates are the continuous spectrum of vascular changes and the separate posterior region of zone II. Plus disease has been the most difficult concept in ROP to teach and standardize since the publication of the “standard” photograph in ICROP.1Committee for the Classification of Retinopathy of PrematurityAn international classification of retinopathy of prematurity.Arch Ophthalmol. 1984; 102: 1130-1134Crossref PubMed Scopus (1121) Google Scholar Since that time, numerous ROP meetings have included discussions about exactly how much dilation and tortuosity and in which vessels are needed to diagnose plus disease. Plus is an important concept because its presence or absence largely determines which eye is to be treated. Research has shown that although individual ROP experts are internally consistent about their diagnosis, there remains wide variability between ROP experts in grading presence or absence of plus disease.5Campbell J.P. Kalpathy-Cramer J. Erdogmus D. et al.Plus disease in retinopathy of prematurity: a continuous spectrum of vascular abnormality as a basis of diagnostic variability.Ophthalmology. 2016; 123: 2338-2344Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar This means at a single site, treatment is being offered at fairly uniform severity threshold, but across clinical centers ROP treatment is being recommended for a variety of ROP severities. In addition, many ROP experts use much of the posterior pole to judge the presence of plus,6Campbell J.P. Ataer-Cansizoglu E. Bolon-Canedo V. et al.Expert diagnosis of plus disease in retinopathy of prematurity from computer-based image analysis.JAMA Ophthalmol. 2016; 134: 651-657Crossref PubMed Scopus (64) Google Scholar rather than just the retinal area immediately around the optic disc as was used in the standard high magnification image in ICROP. The ICROP3 begins to adjust plus disease guidance to match what clinicians have already been doing. The examining ophthalmologist should use the vascular changes throughout zone I to assess presence or absence of plus disease (Fig 1). In addition, plus disease no longer includes signs of advanced disease: vascular engorgement of the iris, poor pupillary dilation, and peripheral retinal vascular engorgement; all are signs of severe disease. Although ICROP3 still includes not-plus, pre-plus, and plus, suggesting there are distinct cut points, the authors remind us that plus is a continuum of vascular change in all 4 quadrants of zone I that needs to be considered. In the future, plus severity will likely be considered on a wider scale and perhaps not rely on a single cut point on the scale for plus/no plus and need for treatment. The ICROP3 begins the transition from a binary yes/no decision to a continuous condition of retinal vascularity in the newborn eye. The second change to note is the division of zone II into posterior and anterior sections. Zone II is where most acute ROP requiring treatment is found, but the natural history and risks differ in this zone, with greater concern for disease in posterior zone II. This addition will allow future study and perhaps even adjustment of treatment indications for posterior disease. The third change is the addition of descriptions of regression and reactivation. Since the advent of anti-VEGF treatment for ROP, the occurrence of reactivation has become a possibility, something that did not occur with ablative therapies such as laser. Reactivated ROP develops between 37 and 60 weeks postmenstrual age, but can occur much later. Planning for monitoring children after anti-VEGF treatment for reactivation is needed and should be arranged before discharge from the hospital. Reactivated ROP does not necessarily behave like pretreatment acute ROP, may progress more rapidly, may be at the leading edge of vascularization or elsewhere in the retina, and may be subtle, yet develop significant fibrosis upon healing. The ICROP3 remains a tool guiding ophthalmoscopic documentation during a physical exam and will take time to learn and record accurately. The recommended changes add substantial detail to the clinical disease description, potentially making it more complicated for the ROP examiner to determine severity and treatment timing. These changes will gradually be incorporated into diagnosis and study data collection, and likely affect the recommended timing and disease severity at treatment, just as CRYO-ROP2Cryotherapy for Retinopathy of Prematurity Cooperative GroupMulticenter trial of cyrotherapy for retinopathy of prematurity.Arch Ophthalmol. 1990; 108: 1408-1416Crossref PubMed Scopus (312) Google Scholar and ETROP7Early Treatment for Retinopathy of Prematurity Cooperative GroupRevised indications for the treatment of retinopathy of prematurity. Results of the early treatment for retinopathy of prematurity randomized trial.Arch Ophthalmol. 2003; 121: 1684-1696Crossref PubMed Scopus (1471) Google Scholar did in earlier decades. While the increased detail of the description of ROP provided by ICROP3 will make it harder for the clinician to completely analyze, these changes seem ideally suited to foster the development of assistive and autonomous diagnostic software for ROP.8Brown J.M. Campbell J.P. Beers A. et al.Automated diagnosis of plus disease in retinopathy of prematurity using deep convolutional neural networks.JAMA Ophthalmol. 2018; 136: 803-810Crossref PubMed Scopus (245) Google Scholar Analytic software could provide a risk estimate of adverse outcomes at each evaluation based on a multi-point scale of vascular severity, zone, and stage over time, as well as incorporating additional factors that influence the clinical course. It is conceivable a computer could examine images of the retina using previous machine learning training to categorize zone I vascular severity in a more objective manner than a clinician can.9Choi R.Y. Brown J.M. Kalpathy-Cramer J. et al.Variability in plus disease identified using a deep learning-based retinopathy of prematurity severity scale.Ophthalmol Retina. 2020; 4: 1016-1021Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar This severity value along with demographic and timing data could be used to predict which eye needs treatment earlier and reduce variability in the diagnosis of plus disease, thereby reducing ROP-associated visual morbidity, without the need to treat many additional babies who would not benefit from treatment. Unfortunately, barriers to adoption of technological assessment of ROP severity remain significant. There is a critical need to adequately fund the development, maintenance, and clinical use of medical software for ROP diagnosis.10Chen M.M. Golding L.P. Nicola G.N. Who will pay for AI?.Radiol Artif Intell. 2021; 3e210030Crossref Scopus (13) Google Scholar In addition, there is the need for photographers skilled with wide-angle ophthalmic cameras who are comfortable with premature infants and who can quickly capture images during every eye examination (Fig 1). The ICROP3 improves our ability to more completely describe and study ROP, but will take time to teach. It may promote incorporation of assistive medical software into our practices to improve the diagnosis and timing of treatment for ROP. International Classification of Retinopathy of Prematurity, Third EditionOphthalmologyVol. 128Issue 10PreviewThe International Classification of Retinopathy of Prematurity is a consensus statement that creates a standard nomenclature for classification of retinopathy of prematurity (ROP). It was initially published in 1984, expanded in 1987, and revisited in 2005. This article presents a third revision, the International Classification of Retinopathy of Prematurity, Third Edition (ICROP3), which is now required because of challenges such as: (1) concerns about subjectivity in critical elements of disease classification; (2) innovations in ophthalmic imaging; (3) novel pharmacologic therapies (e.g., anti–vascular endothelial growth factor agents) with unique regression and reactivation features after treatment compared with ablative therapies; and (4) recognition that patterns of ROP in some regions of the world do not fit neatly into the current classification system. Full-Text PDF