Journal of Pediatric Ophthalmology and Strabismus

Original Article 

ROP Screening Tool Assessment and Validation in a Third-Level Hospital in Argentina: A Pilot Study

Evangelina Esposito, MD, ChM; Erna Knoll, MD; Carla Guantay, MD; Alejandro Gonzalez-Castellanos, MD; Alejandra Miranda, MD; Maria F. Barros Centeno, MD; Martha Gomez Flores, MD; Julio A. Urrets-Zavalia, MD, PhD

Abstract

Purpose:

To evaluate whether a mathematical tool that predicts severe retinopathy of prematurity (ROP) using clinical parameters at 6 weeks of life (ROPScore calculator smartphone application; PABEX Corporation) can be useful to predict severe ROP in a population of premature infants in Argentina.

Methods:

In this retrospective study, data from the clinical records of all premature infants examined between 2012 and 2018 in the ophthalmology department of a public third-level hospital in Córdoba, Argentina, were obtained. ROPScore screening was applied using a Microsoft Excel spreadsheet (Microsoft Corporation). The sensitivity, specificity, and positive (PPV) and negative (NPV) predictive values of the algorithm were analyzed.

Results:

Between 2012 and 2018, a total of 2,894 pre-term infants were examined and 411 met the inclusion criteria, of whom 34% (n = 139) presented some form of ROP and 6% (n = 25) developed severe forms that required treatment. The sensitivity of the algorithm for any ROP and severe ROP was 100%. The PPV and NPV were 35.64% and 100%, respectively, for any ROP and 9.88% and 100% for severe ROP.

Conclusions:

One-time only calculation of the ROPScore algorithm could identify severe cases after validation, reducing the number of screened infants by 38% in infants with a birth weight of 1,500 g or less or a gestational age of 32 weeks or younger.

[J Pediatr Ophthalmol Strabismus. 2021;58(1):55–61.]

Abstract

Purpose:

To evaluate whether a mathematical tool that predicts severe retinopathy of prematurity (ROP) using clinical parameters at 6 weeks of life (ROPScore calculator smartphone application; PABEX Corporation) can be useful to predict severe ROP in a population of premature infants in Argentina.

Methods:

In this retrospective study, data from the clinical records of all premature infants examined between 2012 and 2018 in the ophthalmology department of a public third-level hospital in Córdoba, Argentina, were obtained. ROPScore screening was applied using a Microsoft Excel spreadsheet (Microsoft Corporation). The sensitivity, specificity, and positive (PPV) and negative (NPV) predictive values of the algorithm were analyzed.

Results:

Between 2012 and 2018, a total of 2,894 pre-term infants were examined and 411 met the inclusion criteria, of whom 34% (n = 139) presented some form of ROP and 6% (n = 25) developed severe forms that required treatment. The sensitivity of the algorithm for any ROP and severe ROP was 100%. The PPV and NPV were 35.64% and 100%, respectively, for any ROP and 9.88% and 100% for severe ROP.

Conclusions:

One-time only calculation of the ROPScore algorithm could identify severe cases after validation, reducing the number of screened infants by 38% in infants with a birth weight of 1,500 g or less or a gestational age of 32 weeks or younger.

[J Pediatr Ophthalmol Strabismus. 2021;58(1):55–61.]

Introduction

Retinopathy of prematurity (ROP) is a multifactorial process1 that affects normal retinal vascularization and can cause severe visual disability or blindness even in high-resource settings.2 Early gestational age (GA)3 and uncontrolled oxygen exposure in a preterm infant are the most important factors contributing to ROP pathogenesis.4–6 Other risk factors include low birth weight (BW),7 poor early postnatal growth,7 maternal, prenatal, and perinatal factors, demographics, medical interventions, comorbidities of prematurity, nutrition, and genetic factors.1

Screening programs in high-income countries include infants born with a BW of less than 1,500 g and a GA of less than 32 weeks.8 More mature pre-term infants are only examined if they present additional risk factors. However, it has been shown that in middle- and low-income countries where ROP is still a major cause of childhood blindness, wider criteria are needed to avoid missing opportunities for early diagnosis and treatment.5 As a result, many more premature infants would need to be screened, significantly increasing the number of ophthalmological examinations and follow-up. Due to the fact that some of these infants may not have a true risk for developing severe ROP, it is important to define strategies to safely diminish unnecessary examinations, optimizing resources and minimizing missed cases.9–11

ROPScore12 is a free mathematical tool consisting of a logistic regression equation to calculate the risk per child. ROPScore is calculated one time per infant at 6 weeks of life. It helps to predict severe ROP by assuming a specific cut-off level, using clinical parameters to decrease the number of screened infants and to reduce the frequency of ophthalmological examinations in low-risk infants.13 This process requires the following parameters: BW, GA, weight gain proportional to BW measured at 6 weeks of life (patient's weight measured at 6 weeks of life minus BW, divided by BW), the presence or absence of blood transfusion until 6 weeks of life, and the need for mechanical ventilation.

Comparing the incidence of ROP from population-based studies is difficult6 because the limit of viability relies on the standard of care provided.14 Therefore, current recommendations suggest validating the score for a given population for best results.12,13 This study aimed to evaluate whether ROPScore can be useful to predict severe ROP in a population of premature infants in Argentina.

Patients and Methods

This analytic, observational, retrospective study was approved by the Ethics Committee of the “Ramón Carrillo” Neonatal-Maternal Hospital and performed in accordance with the principles of the Declaration of Helsinki. Informed consent was obtained from the parents of all participating infants.

Data were recorded from clinical files of all premature infants consecutively screened between 2012 and 2018 at the ophthalmology department of a public tertiary referral hospital in Córdoba, Argentina, where screening is performed at a defined moment following our national guidelines.15 In those guidelines, it is recommended to perform screening for ROP in all newborns with a BW of less than 1,500 g and/or a GA of 32 weeks or younger, and in every infant with a GA of between 33 and 36 weeks, with any weight, who presents at least one of the situations identified as risk factors for ROP (eg, oxygen therapy, delayed intrauterine or extrauterine growth, transfusions, sepsis, and early administration of erythropoietin). The first screening is defined according to Table 1.

Screening Protocol

Table 1:

Screening Protocol

It is recommended that children older than 33 weeks of GA at birth who have received oxygen or have aggravating factors be examined within 1 to 2 weeks to confirm whether vascularization of the retina has been completed. All infants in our study were examined by one of the authors (EK) by binocular indirect ophthalmoscopy through a 28.00-diopter lens and with the aid of a scleral depressor for scleral indentation and/or globe rotation.

Preterm infants born with a BW of 1,500 g or less and/or a GA of 32 weeks or younger were born at or admitted from another institution within 24 hours of birth to the “Ramón Carrillo” Maternal-Neonatal Hospital, Córdoba, Argentina. Exclusion criteria were infants with incomplete medical records or who died before reaching 45 weeks of corrected gestational age and infants in whom weight was not obtained at 6 weeks of life.

The ROPScore was calculated once per infant at 6 weeks of life using a Microsoft Excel spreadsheet (Microsoft Corporation) as described by Eckert et al12 using BW in grams, GA in weeks, blood transfusion up to 6 weeks of life, oxygen in mechanical ventilation up to 6 weeks of life, and weight at completed 6 weeks of life.

Agreement between Excel software (Microsoft Corporation) and ROPScore calculator smartphone application (ROP Score 3 for IOS–ROPScore for Android; PABEX Corporation) was assessed.

Results are presented with descriptive and inferential statistics using Microsoft Excel and GraphPad Prism software (GraphPad Corporation). The sensitivity, specificity, and positive (PPV) and negative (NPV) predictive values of the algorithm were analyzed using chi-square and Fisher's exact tests. The trade-off of sensitivity versus specificity was plotted as a receiver operating characteristic curve.

Results

During the study period, 2,894 preterm infants were examined, 26% (n = 762) of whom had a BW of 1,500 g or less or a GA of 32 weeks or younger. Inclusion criteria were met by 411 infants; 34% (n = 139) presented some form of ROP and 6% (n = 25) developed severe forms (type 1 ROP) that required treatment. Three infants had aggressive posterior ROP (AP-ROP), representing 2.15% of any ROP, 12% of severe ROP, and 0.7% of the total cohort. The characteristics of the population are shown in Table 2. In this cohort, 84% (21 of 25) of infants with severe ROP received transfusions and 40% (10 of 25) needed oxygen with mechanical ventilation.

Demographic Characteristics of All 411 Study Patients

Table 2:

Demographic Characteristics of All 411 Study Patients

ROPScore was first applied with a cut-off value of 11 for any ROP and 14.5 for severe ROP, as described in the literature.12

As shown in Table 3, the cut-off value for ROPScore suggested in the literature would miss 28% of severe cases. Therefore, a new cut-off value was calculated. Sensitivity and specificity curves of the algorithm for any ROP and for severe ROP was calculated for our cohort (Figure 1). The new proposed cut-off value for our cohort is 9.15 points for any ROP and 12.05 points for severe ROP. Sensitivity of the algorithm for both any ROP and severe ROP was 100%. The PPV and NPV were 35.64% and 100%, respectively, for any ROP and 9.88% and 100% for severe ROP (Table 4). Applying the new cut-off value would have prevented 38% of infants from screening in this group (Figure 1).

Utility of ROPScorea in Argentina

Table 3:

Utility of ROPScore in Argentina

Receiver operating characteristic (ROC) curves for the detection of any stage retinopathy of prematurity (ROP) (orange) and severe ROP (blue) according to the ROPScore algorithm.

Figure 1.

Receiver operating characteristic (ROC) curves for the detection of any stage retinopathy of prematurity (ROP) (orange) and severe ROP (blue) according to the ROPScore algorithm.

Accuracy of the ROPScorea for Predicting the Development of ROP in a Cohort With a New Cut-off Value

Table 4:

Accuracy of the ROPScore for Predicting the Development of ROP in a Cohort With a New Cut-off Value

The average of screening times per infant was 5 times (range: 3 to 8 times) for infants free of disease, and 7 times (range: 1 to 12 times) for infants with severe ROP before treatment. Mean time to ROP diagnosis was 5 weeks (range: 1 to 14 weeks) and to ROP treatment was 10 weeks (range: 6 to 15 weeks) of postconceptional age, and the AP-ROP group developed the disease between 6 and 9 weeks of life.

Agreement between Excel and ROPScore calculator application was 100%.

Discussion

Annually, 9% (2,088 of 22,992) of all infants born in Cordoba are preterm, and 35% (n = 730) of these are born with a BW of 1,500 g or less,16 in agreement with published data.6 In our 7-year cohort, we found a higher prevalence of ROP than that reported by some centers12,17 but consistent with the prevalence reported by others.6 Additionally, 6% of screened infants had severe ROP. This is consistent with data reported from Brazil by Eckert et al.12 Reported prevalence of any ROP in Latin America ranged from 6.6% to 82%, and type 1 ROP ranged from 1.2% to 23.8%,18 which is consistent with our data. Regarding AP-ROP, a rate of 2.15% was noted. This is less than that reported by Chang,1 but consistent with reports of approximately 3% from Argentina's national registry.19

ROP risk models that consider BW, GA, and postnatal weight gain include WINROP,20 CHOP ROP,21 OMA-ROP,22 CO-ROP,23 and DIGIROP-Birth.24 The success rates of screening reductions in the number of infants screened varies with every strategy, ranging from 10% to 58%.22,23,25 The use of the most assessed ROP predictive models (WINROP and CHOP-ROP) in neonatal intensive care units from middle- and low-income countries is limited13 and they require complex data. OMA-ROP, CO-ROP, and DIGIROP-Birth are simpler but novel models that need further validation.

One of the advantages of ROPScore is that it is an easy-to-use ROP predictive tool that can be routinely applied once per infant at 6 weeks of life by the health care team during screening for ROP. Clinical parameters are used, avoiding an invasive intervention on a preterm infant. In our report, with the validated cut-off ROPScore value customized for our population, a reduction of 38% of the screened infants was achieved.

One shortcoming of the ROPScore is that it is done at 6 weeks of life, which may delay diagnosis of AP-ROP.17 It was recently proposed that ROPScore assessment can be done at 2 or 4 weeks of life, which would allow the score to be set earlier. The algorithm calculation is the same for any week; therefore, validation for each population must be done in the same manner for all of the cohort and set ROPScore limits accordingly. One limitation of our study is that weight at 2 weeks of life was not available because the second week is not an ROP screening week, as defined by our national guidelines.15 Nonetheless, the current literature exclusively contemplates weight at 6 weeks.12,17,26 Additionally, it is important to note that none of our patients with severe ROP were missed using the score at 6 weeks because all severe ROP cases in our cohort developed between 6 and 9 weeks of life. In other centers, diagnosis is being made between 5 and 9 weeks of life and treatment between 8 and 12 weeks of life27 which is consistent with our results. Comparing with other authors,28 it has been found that AP-ROP sometimes starts as a discrete and ill-defined ROP, and after a few weeks develops more obvious characteristics of AP-ROP.

From our ROPScore assessment, we found a lack of precision in the definitions of the parameters of the score, and this could be dangerous. Currently, ROP is followed by the treating team comprising nurses, neonatologists, and ophthalmologists, all of whom manage a presumably validated score for their patients to hypothetically reduce the number of screened infants.

Although ROPScore appears to be a good tool, its application draws our attention to some areas of concern. When applying ROPScore, the GA, BW, and blood transfusions have straightforward definitions and the number is clear. Nonetheless, when defining the completed 6-week weight, the score does not specify if we should consider day 42 or 47 of life. If we consider that premature infants should have an average weight gain of 15 to 20 g/kg/d29 and we calculate hypothetical weight variations at 6 weeks of life, we demonstrate that the final ROPScore is slightly affected. Therefore, it would be interesting to develop prospective studies in this matter, or at least consider this possibility when defining the validated ROPScore cut-off value.

Regarding oxygen in mechanical ventilation as a binary variable, we had trouble defining what to consider mechanical ventilation. In our hospital, infants born with respiratory distress are often exposed to free oxygen in the delivery room and some of them are intubated at least for a few hours. Furthermore, supplementary oxygen can be administered using halo or nasal cannula, and positive pressurized oxygen can be administrated with methods such as time-cycled pressure limited, continuous positive airway pressure (CPAP), and noninvasive ventilation. In this matter, to strictly follow the definitions of the score and after discussion with the neonatologists, we defined “mechanical ventilation” as any variant of intermittent mandatory ventilation or patient-triggered ventilation,30 including noninvasive ventilation, CPAP, and time-cycled pressure limited. This is regardless of the exposure hours, the percentage of hemoglobin that is saturated with oxygen (SpO2, default alarm limits set with the lower limit at 89% and the upper limit at 95%), or the fraction of inspired oxygen (FiO2). It is important to clarify that some other screening algorithms take into account the length of oxygen therapy received by the infant.1 Nevertheless, we state that a more precise definition would add to the accuracy of the score.

ROPScore application must include all severe cases, otherwise we would not be preventing blindness. This caused some concern when analyzing published data. The cut-off point for severe ROP proposed by Eckert et al12 is 14.5 points despite having an ROPScore range for severe ROP between 12.9 and 20.7 points (n = 24), missing 4% of cases with severe ROP.12,25 Similarly, Lucio et al26 proposed a cut-off value of 16.6 points for severe ROP, reporting an ROPScore range for severe ROP between 14.7 and 19.6 points (n = 22). This means severe cases outside of the cut-off value. On the contrary, a multicentric study by Piermarocchi et al17 proposed a cut-off of 15.8 points due to all severe ROP cases scored above this number in that population (n = 44). Differences in the cut-off values may arise because, by inherent nature, populations are different. If we compare data from Brazil12,26 and Argentina, features are more likely to be similar due to location, medical care and facilities, and socioeconomic realities. Data between countries are consistent, with the exception of a lower ROPScore for the entire cohort that shows a difference of more than 3 points (Table 5). In contrast, Italy has a higher income than Argentina, and this could be a logical explanation to account for the differences observed between ROPScore ranges for severe ROP17 and the fact that their cohort had a much younger GA (3 weeks) than in our cohort.

Comparisons Between Current Study and Previous Reports About ROPScorea

Table 5:

Comparisons Between Current Study and Previous Reports About ROPScore

The ROPScore range differs between populations studied.12,17,26 This may be related to the capability of the ophthalmologist to detect and classify ROP, and to a lack of agreement between ophthalmologists.1 Also, it has frequently been reported that inter-ophthalmologist disparity exists when staging ROP.31

Piermarocchi et al17 stressed that most of the infants in their cohort were infants with severe ROP who received transfusions and oxygen through mechanical ventilation. The fact that 84% (n = 21) of our infants with severe ROP received transfusions strongly agrees with the former conclusion but shows more variation regarding mechanical ventilation. Our results in this matter are similar to the work of Arima et al.32 They proposed that the prediction of type 1 ROP is improved by adding late-onset circulatory collapse, CPAP, and the ROPScore model.

Limitations of our study are the retrospective nature and the lack of accuracy in the transfusion date, and whether the transfusion was done at 6 weeks of life or later. The retrospective nature of our study does not allow for the consideration of confounding factors or control groups, which may affect the conclusions of this work. To add to this growing international knowledge, prospective studies on ROPScore application are needed.

The ROPScore algorithm is a screening tool that needs validation for each population. It is a simple and easy tool to apply, with customizable interpretations of results, making this a promising algorithm. In our cohort, a cut-off value for severe ROP of 12 might be accurate to spot severe cases, but careful interpretation of ROPScore is advised and prospective studies are needed. The ROPScore algorithm could reduce the number of ophthalmological examinations necessary to detect ROP by 38% in infants with a BW of 1,500 g or less or a GA of 32 weeks or younger in this cohort. The ROPScore calculator for smartphones is an easy access calculation tool for ROPScore.

References

  1. Chang JW. Risk factor analysis for the development and progression of retinopathy of prematurity. PLoS One. 2019;14(7):e0219934. doi:10.1371/journal.pone.0219934 [CrossRef]
  2. Norman M, Hellström A, Hallberg B, et al. Prevalence of severe visual disability among preterm children with retinopathy of prematurity and association with adherence to best practice guidelines. JAMA Netw Open. 2019;2(1):e186801. doi:10.1001/jamanetworkopen.2018.6801 [CrossRef]
  3. Isaza G, Arora S. Incidence and severity of retinopathy of prematurity in extremely premature infants. Can J Ophthalmol. 2012;47(3):296–300. doi:10.1016/j.jcjo.2012.03.027 [CrossRef]
  4. Early Treatment For Retinopathy Of Prematurity Cooperative Group. Revised indications for the treatment of retinopathy of prematurity: results of the early treatment for retinopathy of prematurity randomized trial. Arch Ophthalmol. 2003;121(12):1684–1694. doi:10.1001/archopht.121.12.1684 [CrossRef]
  5. Gilbert C, Fielder A, Gordillo L, et al. International NO-ROP Group. Characteristics of infants with severe retinopathy of prematurity in countries with low, moderate, and high levels of development: implications for screening programs. Pediatrics. 2005;115(5):e518–e525. doi:10.1542/peds.2004-1180 [CrossRef]
  6. Hellström A, Smith LE, Dammann O. Retinopathy of prematurity. Lancet. 2013;382(9902):1445–1457. doi:10.1016/S0140-6736(13)60178-6 [CrossRef]
  7. Wikstrand MH, Hård AL, Niklasson A, Smith L, Löfqvist C, Hellström A. Maternal and neonatal factors associated with poor early weight gain and later retinopathy of prematurity. Acta Paediatr. 2011;100(12):1528–1533. doi:10.1111/j.1651-2227.2011.02394.x [CrossRef]
  8. Larsson E, Holmström G. Screening for retinopathy of prematurity: evaluation and modification of guidelines. Br J Ophthalmol. 2002;86(12):1399–1402. doi:10.1136/bjo.86.12.1399 [CrossRef]
  9. Gilbert C, Muhit M. Eye conditions and blindness in children: priorities for research, programs, and policy with a focus on childhood cataract. Indian J Ophthalmol. 2012;60(5):451–455. doi:10.4103/0301-4738.100548 [CrossRef]
  10. Urrets-Zavalia JA, Crim N, Knoll EG, et al. Impact of changing oxygenation policies on retinopathy of prematurity in a neonatal unit in Argentina. Br J Ophthalmol. 2012;96(12):1456–1461. doi:10.1136/bjophthalmol-2011-301394 [CrossRef]
  11. Fierson WMAmerican Academy of Pediatrics Section on OphthalmologyAmerican Academy of OphthalmologyAmerican Association for Pediatric Ophthalmology and StrabismusAmerican Association of Certified Orthoptists. Screening examination of premature infants for retinopathy of prematurity. Pediatrics. 2018;142(6):e20183061. doi:10.1542/peds.2018-3061 [CrossRef]
  12. Eckert GU, Fortes Filho JB, Maia M, Procianoy RS. A predictive score for retinopathy of prematurity in very low birth weight preterm infants. Eye (Lond). 2012;26(3):400–406. doi:10.1038/eye.2011.334 [CrossRef]
  13. Binenbaum G. Algorithms for the prediction of retinopathy of prematurity based on postnatal weight gain. Clin Perinatol. 2013;40(2):261–270. doi:10.1016/j.clp.2013.02.004 [CrossRef]
  14. Allen MC, Donohue PK, Dusman AE. The limit of viability—neonatal outcome of infants born at 22 to 25 weeks' gestation. N Engl J Med. 1993;329(22):1597–1601. doi:10.1056/NEJM199311253292201 [CrossRef]
  15. Grupo ROP Argentina, Ministerio de Salud. Guía de Práctica Clínica para la prevención, diagnóstico y tratamiento de la retinopatía del prematuro (ROP). Ministerio de Salud; 2016.
  16. Ministerio de salud Cordoba. Departamento central de estadística. Direccion general de sistemas de informacion. Anuario estadistico 2016. Accessed June 2019. https://datosestadistica.cba.gov.ar/dataset/cfeb194f-c54f-41df-b4b6-1d524d0683eb
  17. Piermarocchi S, Bini S, Martini F, et al. Predictive algorithms for early detection of retinopathy of prematurity. Acta Ophthalmol. 2017;95(2):158–164. doi:10.1111/aos.13117 [CrossRef]
  18. Carrion JZ, Fortes Filho JB, Tartarella MB, Zin A, Jornada ID Jr, . Prevalence of retinopathy of prematurity in Latin America. Clin Ophthalmol. 2011;5:1687–1695.
  19. Lomuto CAE, Avila A, Benítez A, et al. Estadísticas ROP en Argentina 2018. Grupo ROP de Argentina; 2019. http://gruporopargentina.blogspot.com
  20. Wu C, Löfqvist C, Smith LE, VanderVeen DK, Hellström A, Consortium WWINROP Consortium. Importance of early postnatal weight gain for normal retinal angiogenesis in very preterm infants: a multicenter study analyzing weight velocity deviations for the prediction of retinopathy of prematurity. Arch Ophthalmol. 2012;130(8):992–999. doi:10.1001/archophthalmol.2012.243 [CrossRef]
  21. Binenbaum G, Ying GS, Tomlinson LA, Postnatal G. Retinopathy of Prematurity Study G. Validation of the Children's Hospital of Philadelphia Retinopathy of Prematurity (CHOP ROP) Model. JAMA Ophthalmol. 2017;135(8):871–877. doi:10.1001/jamaophthalmol.2017.2295 [CrossRef]
  22. McCauley K, Chundu A, Song H, High R, Suh D. Implementation of a clinical prediction model using daily postnatal weight gain, birth weight, and gestational age to risk stratify ROP. J Pediatr Ophthalmol Strabismus. 2018;55(5):326–334. doi:10.3928/01913913-20180405-02 [CrossRef]
  23. Cao JH, Wagner BD, McCourt EA, et al. The Colorado-Retinopathy of Prematurity Model (CO-ROP): postnatal weight gain screening algorithm. J AAPOS. 2016;20(1):19–24. doi:10.1016/j.jaapos.2015.10.017 [CrossRef]
  24. Pivodic A, Hård AL, Löfqvist C, et al. Individual risk prediction for sight-threatening retinopathy of prematurity using birth characteristics. JAMA Ophthalmol. 2019;138(1):1–9. doi:10.1001/jamaophthalmol.2019.4502 [CrossRef]
  25. Hutchinson AK, Melia M, Yang MB, VanderVeen DK, Wilson LB, Lambert SR. Clinical models and algorithms for the prediction of retinopathy of prematurity: a report by the American Academy of Ophthalmology. Ophthalmology. 2016;123(4):804–816. doi:10.1016/j.ophtha.2015.11.003 [CrossRef]
  26. Lucio KCDV, Bentlin MR, Augusto ACL, et al. The ROPScore as a screening algorithm for predicting retinopathy of prematurity in a Brazilian population. Clinics (São Paulo). 2018;73:e377. doi:10.6061/clinics/2018/e377 [CrossRef]
  27. Lundgren P, Lundberg L, Hellgren G, et al. Aggressive posterior retinopathy of prematurity is associated with multiple infectious episodes and thrombocytopenia. Neonatology. 2017;111(1):79–85. doi:10.1159/000448161 [CrossRef]
  28. Katoch D, Dogra MR, Aggarwal K, et al. Posterior zone I retinopathy of prematurity: spectrum of disease and outcome after laser treatment. Can J Ophthalmol. 2019;54(1):87–93. doi:10.1016/j.jcjo.2018.03.005 [CrossRef]
  29. VanderVeen DK, Martin CR, Mehendale R, Allred EN, Dammann O, Leviton AELGAN Study Investigators. Early nutrition and weight gain in preterm newborns and the risk of retinopathy of prematurity. PLoS One. 2013;8(5):e64325. doi:10.1371/journal.pone.0064325 [CrossRef]
  30. Claure N, Bancalari E. New modes of mechanical ventilation in the preterm newborn: evidence of benefit. Arch Dis Child Fetal Neonatal Ed. 2007;92(6):F508–F512. doi:10.1136/adc.2006.108852 [CrossRef]
  31. Campbell JP, Ryan MC, Lore E, et al. Imaging & Informatics in Retinopathy of Prematurity Research Consortium. Diagnostic discrepancies in retinopathy of prematurity classification. Ophthalmology. 2016;123(8):1795–1801. doi:10.1016/j.ophtha.2016.04.035 [CrossRef]
  32. Arima M, Tsukamoto S, Fujiwara K, Murayama M, Fujikawa K, Sonoda KH. Late-onset circulatory collapse and continuous positive airway pressure are useful predictors of treatment-requiring retinopathy of prematurity: a 9-year retrospective analysis. Sci Rep. 2017;7(1):3904. doi:10.1038/s41598-017-04269-5 [CrossRef]

Screening Protocol

Gestational Age (Weeks)First Screening Examination Date (Postnatal Weeks)
229
238
247
256
265
274
284
294
304
313
322
332

Demographic Characteristics of All 411 Study Patients

CharacteristicTotal CohortPatients With Any Stage ROPPatients With Severe ROP
No. of patients41113925
Mean ± SD BW (g)1,240.12 ± 331.621,051.40 ± 246.63927.80 ± 165.04
Mean ± SD GA (weeks)30.28 ± 2.3229.01 ± 2.2527.92 ± 1.93
Mean ± SD WG at 6 weeks of life (g)697.22 ± 321.82579.99 ± 325.84493.04 ± 240.71
Mean ± SD ROPScorea (range)12.88 ± 2.44 (6.72 to 19.85)14.25 ± 2.16 (9.19 to 19.35)15.50 ± 1.78 (12.10 to 17.89)

Utility of ROPScorea in Argentina

Proposed ROPScore Cut-off ValueCases DetectedCases MissedTotal% of Missed ROP Cases
Any ROP (11)128111397.91
Severe ROP (14.5)1872528

Accuracy of the ROPScorea for Predicting the Development of ROP in a Cohort With a New Cut-off Value

ParameterAny Stage ROP (ROPScore > 9.15)Severe ROP (ROPScore > 12.05)
Sensitivity (%), range100 (97.31 to 100)100 (86.68 to 100)
Specificity (%), range7.72 (5.105 to 1.151)40.93 (36.14 to 1.693)
PPV (%), range35.64 (31.05 to 40.51)9.881 (6.783 to 14.18)
NPV (%), range100 (84.54 to 100)100 (97.63 to 100)

Comparisons Between Current Study and Previous Reports About ROPScorea

GroupArgentina (Current)Brazil (Eckert et al12)Italy (Piermarocchi et al17)Brazil (Lucio et al26)
Total cohort
  No. of patients411474445181
  Mean ± SD BW (g)1,240.12 ± 331.621,217.3 ± 2721,188.8 ± 372.2b1,271 ± 354.6
  Mean ± SD GA (weeks)30.28 ± 2.3230.3 ± 2.231.3 ± 2.3b29.2 ± 2.2
  Mean ± SD WG at 6 weeks of life (g)697.22 ± 321.82637.2 ± 242.1596 ± 248.0
  Mean ± SD ROP-Score (range)12.88 ± 2.44 (6.72 to 19.85)13.6 ± 2.5 (9.1 to 21.6)13.5 ± 3.0 (7.2 to 19.6)
Severe ROP
  No. of patients25244422
  Mean ± SD BW (g)927.80 ± 165.04908.7 ± 232.6up to 1,499763.1 ± 186.8
  Mean ± SD GA (weeks)27.92 ± 1.9327.9 ± 2.225.26 (range: 23 to 28)25.9 ± 1.2
  Mean ± SD WG at 6 weeks of life (g)493.04 ± 240.71411.7 ± 277.4390.7 ± 162.8
  Mean ± SD ROP-Score (range)15.50 ± 1.78 (12.10 to 17.89)17.0 ± 1.8 (12.9 to 20.7)15.8 lowest17.9 ± 1.0 (14.7 to 19.6)
Authors

From the Department of Ophthalmology, Clínica Universitaria Reina Fabiola, Universidad Católica de Córdoba, Córdoba, Argentina (EE, EK, CG, AG-C, AM, MFBC, JAU-Z); and the Department of Ophthalmology, Hospital Materno Neonatal “Ramón Carrillo,” Córdoba, Argentina (EK, MGF).

The authors have no financial or proprietary interest in the materials presented herein.

This research was undertaken as part of the second year of the ChM (Master of Surgery) in Clinical Ophthalmology, Edinburgh Surgery Online (ESO), University of Edinburgh.

The authors thank Dr. Debra Meghan Sanft and Miss Christina Mastromonaco for their dedication to the manuscript preparation.

Correspondence: Evangelina Esposito, MD, ChM, Department of Ophthalmology, Clínica Universitaria Reina Fabiola, Universidad Católica de Cordoba, Jacinto Rios 554, Piso 7, Cordoba 5000, Argentina. Email: gely_e@hotmail.com

Received: April 27, 2020
Accepted: August 05, 2020

10.3928/01913913-20201102-01

Sign up to receive

Journal E-contents