Article Text

Download PDFPDF

Development of a joint set of database parameters for the EU-ROP and Fight Childhood Blindness! ROP Registries
  1. Caroline Catt1,2,
  2. Johanna M Pfeil3,
  3. Daniel Barthelmes2,4,
  4. Glen A Gole5,
  5. Tim U Krohne6,
  6. Wei-Chi Wu7,
  7. Shunji Kusaka8,
  8. Peiquan Zhao9,
  9. Shuan Dai5,
  10. James Elder10,
  11. Matthias Heckmann11,
  12. Jacqueline Stack12,
  13. Gigi Khonyongwa-Fernandez13,
  14. Andreas Stahl3
  1. 1 Department of Ophthalmology, The Children's Hospital at Westmead, Westmead, New South Wales, Australia
  2. 2 The Save Sight Institute, The University of Sydney, Sydney, New South Wales, Australia
  3. 3 Department of Ophthalmology, University Medicine Greifswald, Greifswald, Mecklenburg-Vorpommern, Germany
  4. 4 Department of Ophthalmology, University Hospital Zurich, Zurich, Zürich, Switzerland
  5. 5 Department of Ophthalmology, Queensland Children's Hospital, Brisbane, Queensland, Australia
  6. 6 Department of Ophthalmology, University of Cologne, Koln, Nordrhein-Westfalen, Germany
  7. 7 Department of Ophthalmology, Chang Gung Memorial Hospital, Kweishan, Taiwan
  8. 8 Department of Ophthalmology, Kindai University Faculty of Medicine Hospital, Osakasayama, Japan
  9. 9 Department of Ophthalmology, Xinhua Hospital Affiliated to Shanghai Jiaotong University School of Medicine, Shanghai, China
  10. 10 Department of Ophthalmology, Royal Children's Hospital, Parkville, Victoria, Australia
  11. 11 Department of Neonatology and Pediatric Intensive Care, University Medicine Greifswald, Greifswald, Mecklenburg-Vorpommern, Germany
  12. 12 Neonatal Intensive Care Unit, Liverpool Hospital, Liverpool, New South Wales, Australia
  13. 13 Families Blossoming LLC, The Health Foundation, London, London, UK
  1. Correspondence to Caroline Catt, The Eye Clinic, The Children's Hospital at Westmead, Westmead, NSW 2145, Australia; carolinecatt{at}


Background/Aims The incidence of retinopathy of prematurity (ROP) is increasing and treatment options are expanding, often without accompanying safety data. We aimed to define a minimal, patient-centred data set that is feasible to collect in clinical practice and can be used collaboratively to track and compare outcomes of ROP treatment with a view to improving patient outcomes.

Methods A multinational group of clinicians and a patient representative with expertise in ROP and registry development collaborated to develop a data set that focused on real-world parameters and outcomes that were patient centred, minimal and feasible to collect in routine clinical practice.

Results For babies receiving ROP treatment, we recommend patient demographics, systemic comorbidities, ROP status, treatment details, ophthalmic and systemic complications of treatment, ophthalmic and neurodevelopmental outcomes at initial treatment, any episodes of retreatment and follow-up examinations in the short and long-term to be collected for use in ROP studies, registries and routine clinical practice.

Conclusions We recommend these parameters to be used in registries and future studies of ROP treatment, to reduce the variation seen in previous reports and allow meaningful assessments and comparisons. They form the basis of the EU-ROP and the Fight Childhood Blindness! ROP Registries.

  • Child health (paediatrics)
  • Retina
  • Treatment Lasers
  • Treatment Surgery
  • Vision

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • The incidence of retinopathy of prematurity (ROP) is increasing, and treatment options are expanding without accompanying safety data. The ability to undertake collaborative studies, pool data and make meaningful comparisons across studies is limited by the absence of a standardised approach to which parameters should be collected on babies treated for ROP.


  • This study provides a standardised set of parameters to guide data collection in ROP trials, registries and routine clinical practice.


  • Adoption of these parameters will allow data to be collated and compared across the world. This is particularly important for treated ROP because it is uncommon but increasing in incidence. Improved understanding of treatment, practice patterns and outcomes may lead to the development of international practice guidelines and most importantly, drive improved patient outcomes.


Retinopathy of prematurity (ROP) is a vision-threatening disease associated with abnormal vascular proliferation in the developing retina of premature babies. Each year, more than 30 000 babies develop vision impairment due to this condition1 and it has been identified as a leading cause of childhood blindness by the WHO Vision 2020 Programme.2 The incidence of ROP is increasing in both developed and developing countries due to the improved survival rates of preterm infants.3–5

The treatment of ROP is an evolving landscape. The evidence generated by randomised, controlled trials (RCTs) supporting the destruction of peripheral avascular retina by either cryotherapy6 or confluent laser photocoagulation7 underwent a paradigm shift after the results of the Bevacizumab Eliminates the Angiogenic Threat of ROP (BEAT-ROP) study were published in 2011. Despite some study limitations, BEAT-ROP raised the possibility that anti-vascular endothelial growth factors (anti-VEGF) may have advantages over laser treatment for zone 1 disease.8 More recently, the ranibizumab versus laser therapy for the treatment of very low birth weight infants with ROP (RAINBOW) study concluded intravitreal ranibizumab (0.2 mg) might be superior to conventional laser therapy, with fewer eyes having unfavourable structural outcomes, and with an acceptable safety profile 24 weeks after treatment.9 Based on the results from RAINBOW, ranibizumab was approved for the treatment of ROP by the European Medicine Agency and Pharmaceuticals and Medical Devices Agency of Japan in 2019 and by the Australian Therapeutic Goods Administration in 2022. Results for aflibercept have also recently been published from a phase 3 trial showing success rates of 85.5%10 and this medication was subsequently approved for use in Europe and Japan in 2022 and the USA in 2023.

Despite the increasing amount of data from controlled clinical trials, several unanswered questions relating to the use of anti-VEGF in preterm infants remain. The ideal medication, dose and ocular efficacy has not been clearly established.9 VEGF plays a key role in angiogenesis during the foetal and neonatal periods and at present, the safety of anti-VEGF agents in preterm infants remains uncertain, especially with drugs suppressing systemic VEGF levels.11 Some of these questions are difficult to assess by RCTs because of the very large sample size required to show differences between treatment groups, as ROP requiring treatment is a rare disease. ROP has been described as a ‘biomarker’ for pathological processes that are also occurring in the developing brain12 and an association between ROP and smaller unmyelinated white matter and cerebellar volumes has been shown.13 It is challenging, if not impossible, to determine by RCT whether a neurodevelopmental outcome is due to treatment for ROP or whether impaired outcomes are an expression of a common underlying disease process—that is, one which causes both ROP and neurodevelopmental compromise.

RCT data may also be limited by a selection bias of the study populations,14 which makes it difficult to generalise findings to more inclusive patient populations. RCT study protocols also differ from the realities of what is possible in real-world clinical practice.15 While they remain the most powerful tool for developing scientific evidence around therapeutic interventions, our increasing understanding of RCT limitations has encouraged the move towards using real-world evidence, that is, evidence arising from data collected in clinical care settings as opposed to academic research environments, as a means of complementing scientific knowledge.15

Clinical Registries have been used in ophthalmology for decades. They are a source of real-world data and are particularly useful for rare diseases, such as ROP. Sweden (SWEDROP) and Germany ( ROP registry) are two countries that have collected and published larger scale longitudinal data from their national Registries. The German ROP registry has recently been transformed into the EU-ROP registry, now allowing multinational data entry from countries across geographical Europe and the launch of the Fight Childhood Blindness! ROP Registry is imminent, also with multinational input. The item list used in EU-ROP was codeveloped together with the Fight Childhood Blindness! ROP group and is the topic of this paper.

Materials and methods

An international Core Working Group (Core Working Group members—Caroline Catt, Johanna M Pfeil, Daniel Barthelmes, Andreas Stahl) of clinicians and ROP experts was formed following communication about mutual interest in developing ROP registries. The Core Working Group members had expertise across ROP screening and treatment, Vitreoretinal surgery, Paediatric ophthalmology and Ophthalmological Registries. The Group met multiple times via video conference to develop a comprehensive data set encompassing patient demographics, accompanying systemic medical conditions, ocular findings and treatment parameters at initial treatment and any episodes of retreatment, systemic and ocular complications of treatment, short-term ocular outcomes and long-term ocular and systemic outcomes. The parameters used in the German ROP registry served as a basis for this endeavour and were adapted according to experiences from the German registry as well as the expertise of the core working group.

In a second stage, the working group was expanded to include ophthalmologists from other countries and regions to ensure international representation and expertise in ROP management within the group as well and additional expertise in vitreoretinal surgery and paediatric ophthalmology. We also included neonatologists and a patient representative to cover most facets of babies affected by ROP requiring treatment. This Expanded Working Group (Expanded Group members—Glen Gole, Tim Krohne, Matthias Heckmann, Jacqueline Stack, Gigi Khonyongwa-Fernandez—Johanna M Pfeil reported their comments back to the group) reviewed and refined the initial data set via video teleconference meetings and email discussion to formulate a minimum data set.

In a third stage, for the Australian-based Fight Childhood Blindness! ROP Registry, the data set was reviewed and modified by a Final Group (Final Group members for Fight Childhood Blindness! ROP Registry—Wei-Chi Wu, Shunji Kusaka, Peiquan Zhao, Shuan Dai, James Elder) of international ophthalmologists with expertise in ROP management. Consensus decisions on the final Fight Childhood Blindness! ROP data set were made during video teleconferences, email discussions and electronic survey. Similarly, the EU-ROP data set has undergone slight modifications, since the second stage, particularly to accommodate the recently published International Classification of Retinopathy of Prematurity, Third Edition (ICROP3) classifications.16

At all stages, the groups were steered to create a data set that focused on parameters and outcomes that were patient-centred, minimal and feasible to measure and document in routine clinical practice.


A summary of the recommendations made by the group is detailed within box 1, tables 1–6 with further details below. In addition to the parameters listed in each table, we also recommend documenting whether any clinical imaging has been performed and its type (wide-angle digital photography, fluorescein angiography, optical coherence tomography, or other).

Box 1

Basic demographics to be collected at the time of initial treatment

Gestational age.

Birth weight.



Whether a preventive treatment was given for retinopathy of prematurity.

Postmenstrual age at initial treatment (calculated from gestational age, date of birth and date of treatment).

Weight at initial treatment.

Table 1

Clinical characteristics at initial treatment of retinopathy of prematurity (ROP)

Table 2

Initial treatment parameters

Table 3

Post-treatment details (recorded at each postoperative follow-up examination)

Table 4

Retreatment parameters

Table 5

Ophthalmic outcomes to be recorded at each long-term follow-up visit

Table 6

Long-term neurodevelopmental outcomes to be recorded at each follow-up visit

The recommended demographic parameters to be collected at initial treatment are listed in box 1. They include gestational age and weight at birth, sex, plurality, age, and weight at initial treatment and whether any treatment was given for the purposes of preventing ROP.

Table 1 lists the recommended clinical characteristics to be collected at the time of initial treatment for ROP at the documenting hospital. A large number of potential clinical characteristics were considered for inclusion at baseline (initial treatment) and these were ultimately refined to four parameters relating to ROP status (ROP stage, ROP zone, plus disease and extent of any neovascularisation present) and five relating to neonatal comorbidities and treatment (supplementary oxygen, ventilatory support, sepsis, necrotising enterocolitis and intraventricular haemorrhage in both registries, plus porencephaly and periventricular leukomalacia in EU-ROP) prior to initial treatment. We define ventilatory support as any invasive or non-invasive method/technique/machine that supports respiratory ventilation, including ventilation via endotracheal intubation, continuous or bilevel positive airway pressure. It does not include high flow nasal prongs or low flow nasal prongs. If there have been previous episodes of treatment at an external hospital, these can be recorded at the time of initial treatment at the documenting site.

Table 2 details the recommended treatment parameters to be documented. Anaesthetic details allow the distinction between general anaesthesia, and other types of anaesthesia, and also whether intubation associated with general anaesthesia was pre-existing or new for the purposes of providing ROP treatment. Treatment modes for inclusion were selected for being the current commonly used treatments for ROP, namely retinal laser photocoagulation, various intravitreal anti-VEGF drugs, cryotherapy/cryocoagulation and/or different vitreoretinal surgical procedures.

Recommended parameters to be collected at post-treatment examinations are listed in table 3. Whether or not the examination was performed under sedation or general anaesthesia is documented. ROP status is again defined and documented according to ICROP3. Answer options for ocular complications of treatment cover the more common adverse events following ROP treatment by the treatment modes listed in table 2.

Date of hospital discharge and weight at that time are also documented.

Table 4 lists the parameters to collect at the time of any episodes of retreatment. Demographic details are minimised to the date of, and weight at retreatment. ROP status at retreatment is again defined and documented according to ICROP3 as stage, extent of neovascularisation, zone and plus disease. Neonatal status at retreatment aims to document significant systemic events prior to retreatment, including whether there was any interruption to planned follow-up. Retreatment parameters mirror those documented for initial treatment.

The long-term ophthalmic outcome parameters are listed in table 5. The decision regarding when to start recording the long-term outcome parameters (listed in table 5) rather than the post-treatment parameters (listed in table 3) is ultimately up to the treating clinician and may be influenced by local factors such as the transfer of care from one clinician or service provider to another at hospital discharge. This variation across sites and physicians is an integral part and reflects real-world practice. As a general recommendation, we suggest changing to the long-term outcome parameters once the retina is fully vascularised or the treating clinician has a very low expectation of ROP recurrence. This would often be accompanied by a significant extension of the ophthalmic review interval to months, rather than weeks. The recommended parameters include anatomical findings including acknowledged complications of ROP and its treatment, and whether any surgical interventions have occurred since the last examination. The best-corrected visual acuity (BCVA) for each eye and the method by which it was measured is recommended, as is the spectacle prescription if it was used to achieve the BCVA. Other recommended parameters include refraction measured under cycloplegia, the presence of a constant or visually significant strabismus, nystagmus and which ophthalmic non-surgical treatments if any, the child is receiving.

The long-term neurodevelopmental outcomes to be recorded at each long-term follow-up visit are detailed in table 6. We suggest recording whether the child can walk and hear at each visit. Once the child is over 2 years, record if they have cerebral palsy. Many children born prematurely remain in the care of paediatric clinics after hospital discharge and undergo standardised psychometric assessment tests at specified ages. For example, babies born in a neonatal intensive care unit in Australia and New Zealand with a gestational age of less than 28 weeks or with a birth weight less than 1000 g are required to have a Bayley Scale of Infant Development assessment tool at the ages of 12 months (corrected) and 24 months (corrected). The results of standardised psychometric assessments regularly done in the respective countries should be documented together with ophthalmologic outcomes. We recommend documenting whether the child is a recipient of low vision support, after the age of 5 years.


To date, there is no standardised approach to collecting information on babies treated for ROP and consequently, there is significant variation in the parameters in existing registries and influential clinical trials (figure 1). With the incidence of ROP increasing, treatment options expanding (often without accompanying long-term safety data) and no standardised approach to which real-world practice parameters should be recorded for ROP treatment and outcomes, we formed a Working Group to develop an international standardised set of registry parameters designed to describe patient demographics and characteristics, treatment parameters and structural, functional and neurodevelopmental outcomes in the short and long-term for babies treated for ROP. The goal is for these parameters to be used to develop ROP registries and as a guide for routine clinical data collection by clinicians who manage babies treated for ROP. Hopefully, these standardised parameters will be used in future studies of ROP treatment and thus reduce the variation seen in previous reports. Ongoing consideration and collaboration will be necessary to ensure the data set remains up to date with developments in clinical care, ROP knowledge and classification and to ensure it reflects the experience and opinions of other research groups. The development of two separate registries (EU-ROP and Fight Childhood Blindness! ROP) using common parameters reflects differences in regulations geographically but still allows collaboration and pooling of deidentified aggregated data.

Figure 1

Parameters reported as a percentage, by existing retinopathy of prematurity databases and large clinical trials: (A) demographic parameters; (B) treatment details and complications and (C) treatment outcomes. Databases and clinical trials included are CRYO-ROP,31 ETROP,7 BEAT-ROP,8 RAINBOW,9 SWEDROP38 and German ROP Registry.39

Clinical registry parameters must strike a balance between having extensive data sets that allow for complex analyses, the availability of test results, the quality of data entered and the costs of collecting data in routine clinical practice. Most clinicians are unlikely to enter information into a registry if the data are not easily available to them or if it takes a long time to do so. Like other registries, our data set represents a compromise—a well-balanced one—taking into account the several aspects mentioned. The development of our list of parameters was by consensus, guided by clinical experience and evidence arising from the existing literature and based on the already existing data set from the German ROP registry.

The basic demographic parameters (box 1) were selected based on their contribution to the risk profile of developing severe ROP. Gestational age and birth weight are the two strongest known risk factors for the development of ROP17 with a higher risk for babies small for their gestational age.18 Despite the conflicting evidence around their contribution to the risk of ROP,17 we also include sex and plurality which were deemed important to clinicians and parents, and relatively easy to collect in clinical practice and because male sex is an independent risk factor for poor outcomes in babies born prematurely.19 Including weight at initial treatment allows calculation of weight gain: poor postnatal weight gain is also a risk factor for developing ROP, including severe ROP17 20 and is a general risk factor for poor outcomes in preterm infants.21 We are fully aware that weight gain is not linear in the period between birth and a potential treatment for ROP, but it is a proxy for weight development during that time and it is relatively easy to document. Whether or not a preventive treatment was given for ROP was included, given the increasing interest in such treatments and the fact that prospective clinical trials involving preventive treatments are being planned by other research groups.

The Clinical Characteristics at initial treatment (table 1) are divided into ROP status at initial treatment and neonatal comorbidities. Neonatal comorbidities were selected based on their contribution to the risk for ROP and adjusted to what was deemed feasible to collect by an ophthalmologist, at the time of initial treatment. Oxygen supplementation is a major risk factor for ROP17 and the need for mechanical ventilation22 and its duration is a known independent risk factor for the development of ROP.17 23 It was decided to use the broad categories of more than 5 days, less than or equal to 5 days, or none, with regard to duration of oxygen supplementation, because it was deemed not feasible for an ophthalmologist to determine and record the total number of days, with accuracy. In addition, as most premature babies receive some form of resuscitation in the delivery room, it was felt recording whether any oxygen or ventilatory support was given would be non-discriminatory. Both the EU-ROP and Fight Childhood Blindness! ROP Registries require each documenting hospital to report their local oxygen saturation target ranges as lower ranges are associated with less ROP.24 Sepsis, necrotising enterocolitis and intraventricular haemorrhage are all risk factors for the development of severe ROP.17 25 Intraventricular haemorrhage is an important complication of prematurity which can result in neurodevelopmental impairment.26 Premature babies with periventricular leukomalacia are at increased risk of cerebral palsy, and its cystic form (porencephaly) may be a risk factor for vision and hearing impairment, compared with children born at term.27

Treatment parameters (both for initial as well as retreatments) include the type of anaesthetic used. One of the stated benefits of intravitreal anti-VEGF injection over laser is that it can be done under topical anaesthesia while laser often requires a general anaesthesia—which many parents like to avoid because of the risk of prolonged reintubation and concerns about possible effects on neurodevelopment.28 However, only the BEAT-ROP study specified the anaesthesia type used in their study protocol (oral, intramuscular or intravenous sedation accompanied by topical anaesthesia) while others, such as the ETROP group state that intraoperative pain control was provided according to ‘standard practices’ at each centre.29 We therefore argue it is important to document the type of anaesthesia used for each treatment.

A disadvantage of anti-VEGF treatment is the frequent examinations that are recommended over a long time to monitor for disease recurrence after treatment.30 Understanding this, we were careful to keep the data points required for postprocedure examinations to an absolute minimum. Within the context of a registry, the number of entries informs us of the number of postprocedure examinations undertaken, so long as the data entry is complete. Whether or not any sedation or anaesthesia was used for the examination helps us understand postprocedure examination practice patterns. Including the ROP status advises regression and recurrence patterns and documenting known complications is important with regards to interpreting treatment outcomes.

Monitoring disease and treatment outcomes is a critical role of clinical registries. The long-term ophthalmic outcomes (table 5) include anatomical findings and ocular comorbidities such as strabismus, abnormal eye movements, refractive error and the need for surgical and non-surgical interventions. Ten-year data from the CRYO-ROP study indicate visual acuity outcomes are worse in children with ROP compared with those without ROP31 and children treated with laser rather than intravitreal bevacizumab have a greater risk of bilateral vision impairment.32 Another study reviewing visual acuity outcomes in New Zealand over a 22 year period found visual impairment secondary to ROP declined over the course of the study period,33 which ended just 1 year after the publication of the BEAT-ROP study. Visual acuity outcome is possibly the single most interesting and important piece of information for patients and their families with regard to the long-term outcome of ROP treatment.

The long-term neurodevelopmental parameters (table 6) include walking and hearing, which are easy for an ophthalmologist to document, and whether there is a diagnosis of cerebral palsy, which carers can usually report. Receipt of vision support was felt to be important for patient advocacy reasons. Scores from standardised psychometric tests give an objective measurement of neurodevelopmental outcomes. The ophthalmologist will likely need to source this information as a written report from the parent or paediatrician. Despite this hurdle, we feel it is a critical registry parameter. At present, the studies looking at neurodevelopmental outcomes of severe ROP are small, predominantly retrospective with relatively short-term follow-up and have contradictory findings.34–37

In conclusion, we present a list of parameters that we hope will help guide the development of aligned ROP registries around the world. The list of parameters has been developed by a multinational group with expertise in ROP and registries, are minimal, patient-centred and practical to collect in routine clinical practice. They form the basis of the EU-ROP and Fight Childhood Blindness! ROP Registries. Collecting standardised parameters will allow data to be collated and compared across the world, with the goal of developing international practice guidelines and most importantly, drive improved patient outcomes.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Ethics statements

Patient consent for publication

Ethics approval

Not applicable.


The authors also wish to acknowledge the contributions of Dr JNJH, who provided a valuable opinion on the parameters during their development and Professor MG for his guidance to the corresponding author on the study process.



  • Contributors The following authors contributed to the study in the ways listed, and met all the requirements of authorship according to the ICMJE recommendations. CC: study concept and design, acquisition and analysis of results, manuscript drafting, editing and approval prior to submission and submission, and guarantor. JMP, DB and AS: study concept and design, acquisition and analysis of results, manuscript drafting, editing and approval prior to submission. GAG, W-CW, SK, PZ, SD, JE: study design, acquisition and analysis of results, manuscript editing and approval prior to submission. TUK: acquisition and analysis of results, manuscript editing and approval prior to submission. MH, JS, GK-F: acquisition of results, manuscript editing and approval prior to submission. The authors also wish to acknowledge the contributions of Dr Jesús Noel Jaurietta Hinojos, who provided a valuable opinion on the parameters during their development and Professor Mark Gillies for his guidance to the corresponding author on the study process.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests CC: previous Scientific Advisory Board membership—Novartis, Aspen. JMP: speaker—Novartis. DB: Novartis—past funding for research for the institution; Bayer—past funding for research, consultancy, paid to institution; Alcon—current, advisor, paid to author. TUK: recipient of lecture fees and consultant fees from Alimera Sciences and Roche; recipient of lecture fees from Allergan; recipient of lecture fees and consultant fees and financial support of research projects from Bayer and Novartis; recipient of lecture fees from Heidelberg Engineering. GAG, W-CW, SK, PZ, SD, JE, MH, JS, GK-F: none declared. AS: speaker: Bayer, Novartis; scientific advisory boards: Bayer, Novartis; research grants: Bayer and Novartis; clinical trials: Bayer and Novartis. The EU-ROP project is currently financed by extramural grants from the University Medicine Greifswald, solicited through the EU-ROP principal investigator from Novartis Pharma AG and Bayer Vital AG. Companies supporting the registry do not have any influence on the content or the design of the registry.The Fight Childhood Blindness! ROP Registry is a Save Sight Registry which receives funding from research and educational grants, and private donations, on a strictly non-binding basis.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles

  • Highlights from this issue
    Frank Larkin