Incidence, severity and risk factors of retinopathy of prematurity in a tertiary care centre in Kerala

INTRODUCTION

Retinopathy of prematurity (ROP) is a vaso-proliferative retinal disorder that increases in incidence with decreasing gestational age (GA) and caused by multiple factors. The mechanism of injury postulated are vasoconstriction, increase in level of vasogenic factors like vascular endothelial growth factor and compensatory neovascularization leading to severe extra-retinal fibro-vascular proliferation, scarring and retinal detachment. It is said to be one of the leading causes of preventable childhood blindness in India.¹

Risk factors are associated with ROP include prematurity, low birth weight, prolonged oxygen exposure, severity of neonatal illnesses, sepsis, acidosis, shock, hypoxia, prolonged ventilator support, need for blood transfusion, intra-ventricular hemorrhage, anemia, high ambient light and vitamin E deficiency. Breast feeding and history of maternal preeclampsia are said to be protective.^1,2

According to a study at Pune by Chaudhari et al the incidence of ROP was overall 22.3% and 9% had blindness due to retinal detachment.³ The incidence of ROP according to a study by Rekha et al was overall 46%, with a highest incidence of 73.3% among newborn with birth weight less than 1000 gm.⁴

The incidence of ROP has increased dramatically due to better survival rate of LBW and preterm babies availing modern neonatal facilities and neonatal care. Lack of manpower, lack of awareness and absence of good screening program increases the gravity of the problem.⁵

International classification of ROP (ICROP) is widely accepted which classifies ROP into stages 1 to 5, plus disease, extend, threshold and pre-threshold disease.⁶ ROP predispose to high myopia, amblyopia, glaucoma, refractive errors, astigmatism, strabismus and retinal detachment.^1,5,7

ROP is developing as one of the major causes of blindness in developing countries.⁸ The studies show that lower gestational age and low birth weight babies born in developing countries has higher incidence of ROP compared to that of developed countries. This may be due to lack awareness and lesser neonatal care facilities.⁹ Developed countries with enough financial support and highly equipped neonatal units with expert hands reduced the prevalence of ROP to 5-8% compared to the higher prevalence of around 30 % in developing countries.¹⁰

Most of the risk factors associated with ROP arise in the neonatal intensive care unit (NICU) and are avoidable. The incidence of ROP can be decreased by monitoring the risk factors, timely screening, regular follow up and appropriate interventions.⁵ As there are very few studies regarding incidence and risk factors of ROP in south India especially from Kerala, this study was proposed.

The primary objective of the study was to find out the incidence and severity of ROP in neonates. Secondary objectives were to find out the maternal and neonatal risk factors associated with development of ROP.

METHODS

This was a hospital-based COHORT study conducted in medical trust hospital Kochi a tertiary care center in Kerala among the neonates admitted in the department of pediatrics and neonatology during the period of January 2018 to August 2021.

Inclusion criteria

Children with birth weight less than 2000 gm/ gestational age less than 34 weeks/ gestational age between 34 to 36 weeks but with risk factors such as: a) cardio-respiratory support, b) prolonged oxygen therapy, c) RDS d) bronchopulmonary dysplasia e) fetal hemorrhage, f) blood transfusion, g) neonatal sepsis, h) exchange transfusion, i) intraventricular hemorrhage, j) apnea k) Poor postnatal weight gain and infants with an unstable clinical course who are at high risk (as determined by the neonatologist or pediatrician) were included in the study.

Exclusion criteria

All children with birth weight more than 2000 grams and uneventful neonatal period, children with major congenital malformations, and suspected chromosomal anomalies and neonates who died before full vascularisation or those who lost follow up were excluded from the study.

The 56 children were identified. Data was collected using a standardized data collection chart, and analysed. The results analyzed by descriptive statistics; logistic regression analysis was used to analyze risk factors related to ROP. A p<0.05 was set as statistically significant.

Data collection

The study was conducted among neonates born, referred or attending the outpatient department of neonatology and paediatric department of a tertiary care hospital at Kochi. The subjects were selected according to the inclusion criteria.

Baseline data was collected for each baby regarding date of birth, sex, single or multiple pregnancies, intrauterine growth retardation and other antenatal details. Clinical assessments and lab investigations for identifying the risk factors were collected as mentioned above. Data was collected using a standardized data collection chart, and analysed.

Sample size calculation

Based on previous study on ROP by Chaudhari et al it is observed that incidence of ROP in India is 22.3% and the minimum sample size required is 52.³

Sample size is determined by formula

z=Co-efficient of significance (2.576), α=Level of significance (1.0%), p=Incidence of ROP (22.3%), q=1-p (77.7%), d=Desired precision (15.0%)

The minimum sample size worked out using the above formula is 52.

Statistical methods

SPSS version 20 was used for statistical analysis. Qualitative (Categorical)variables were expressed as percentage. Quantitative (continues) variables represented by mean and standard deviation. Chi-square test &Fisher exact test were used to find out the risk factors associated with ROP. A p value less than 0.05 was taken as statistically significant.

Relevant institutional review board and ethics committee provided approval for the study (Reference no-DNB/0411/2021), at the medical trust hospital, Ernakulam.

RESULTS

A total of 56 neonates who satisfied the inclusion criteria were identified during the study period. Out of the 56 newborns studied, 12 babies developed ROP, and the incidence was found to be 21.4% (12/56). 41.6% (5/12) each developed stage 1 and stage 2 disease, 16.66% (2/12) developed stage 3 disease whereas none developed stage 4 or stage 5 disease. Maternal risk factors including PIH, GDM, antenatal steroid administration, antepartum hemorrhage, delivery method and multiple gestation were studied and no statistically significant association was observed. Lower birth weight was significantly associated with increase in incidence of ROP with highest incidence in the babies with extremely low birth weight. Incidence of ROP was higher in males but there was no significant association between gender and ROP. Incidence of ROP was found higher in lower gestational age group with highest in the extreme preterm and statistically significant association was found. Surfactant administration, longer duration of oxygen therapy, anemia, need for blood transfusion, sepsis and phototherapy were significantly associated with development of ROP.

DISCUSSION

Characteristics of study population

This study was conducted in a tertiary care hospital in Kerala, India, among newborn and infants born in our hospital from the period of 2018 to 2021. During the study period, 56 infants were identified of whom 50% (28/56) were males and 50% (28/56) were females. The study included 4 extremes preterm, 18 very preterm, 7 moderate preterm and 27 late preterm babies (Table 1). The study population, predominantly consisted of low birth weight 69.9% (39/56) followed by very low birth weight 10.7% (6/56) and extremely low birth weight 19.6% (11/56) (Figure 1).

Incidence of ROP

The incidence of ROP in our study is 21.4%. This incidence is comparable to other studies in India. A lower incidence of ROP (13.7%) was found in a study from Ghana.¹¹ A prospective cohort study from Pune by Choudhari among 552 infants was found to have an incidence of 22.3%.³ Takare et al in a prospective observational study from Aurangabad showed an incidence of 27.73%. ¹² But in a study involving 100 babies done in a tertiary care hospital in South India, Rekha S et al found a higher incidence of ROP (46%). ⁴

In a retrospective cohort study from Brazil conducted on babies born before 37 weeks, Freitas found an incidence of 33.9%.¹³ In a study of 308 newborns of gestational age less than 34 weeks, Verma et al reported an incidence of 23.4%.¹⁴

Severity of ROP

Among the 12 babies who developed ROP, only three of them required treatment (3/12- 25%). None of them required surgical correction. Stage 1 and stage 2 disease was found in 5 infants each (5/12-41.67%) and stage 3 disease in 2 infants (2/12-16.66%). None of the infants developed stage 4 or stage 5 disease (Figure 2). A study by Patel SS et al also showed similar distribution with higher percentage of neonates developing stage 1 disease (43.5%) followed by stage 2(30.4%) and stage 3 (7.3%).¹ In a study in Africa, Hakeem et al found that 54.5% developed stage 1 disease, 27.3% developed stage 2 disease and 18.2% developed stage 3 disease.¹⁵ A prospective study by Fielder found that all ROP of stage 1 and 2 underwent complete resolution while stage 3 and 4 developed cicatricial sequel.¹⁶ A study from Mangalore by Shetty found that 50% of neonates with ROP developed stage 1 ROP and 50% developed stage 2 disease and 33.3% of both stages regressed spontaneously.¹⁷

Association between maternal factors and ROP

Pregnancy induced hypertension (PIH) and ROP

In our study out of 16 babies born to mother with PIH 31.2% (5/16) babies developed ROP whereas 17.5% (7/40) babies born to mothers without PIH developed ROP. There was no association between PIH and ROP (Table 2). PIH and the risk of ROP in preterm infants were studied by Alshaikh, which found no significant correlation.¹⁸ Yau et al in their study also did not find a significant correlation between PIH and the development of ROP.¹⁹ Gebesce et al. in their study showed higher incidence of ROP in babies born to mothers with PIH, which is contradictory to our outcome.²⁰

Gestational diabetes mellitus (GDM) and ROP

In our study population 15 babies were born to mothers with GDM. Among them only 6.7% (1/15) developed ROP while 93.3% (14/15) babies did not and no significant correlation was found (Table 2). Kavurt et al in their study in 2014 including 495 patients also did not find any correlation between the incidence of ROP and maternal risk factors like GDM and PIH. ²¹

Antenatal steroid and ROP

In our study 47 mothers received antenatal steroid, out of which 23.4% (11/47) developed ROP.9 mothers did not receive antenatal steroid out of which 11.1% (1/9) developed ROP. Antenatal steroid was also not found to have any association with the development of ROP (Table 2). Kumar et al also showed no association with ROP and antennal steroid.⁵

Antepartum hemorrage and ROP

The 14.3% (1/7) of neonates with maternal antepartum hemorrhage developed ROP compared to 22.4% (11/49) without antepartum hemorrhage. It was also statistically not significant (Table 2). Shah conducted a study in Singapore which also did not find any significant association between antepartum hemorrage and development of ROP.²²

Delivery methods and ROP

In our study, we found that 41.7% (5/12) babies born by normal vaginal delivery developed ROP while 15.91% (7/44) babies born by LSCS developed ROP. There was no significant association (Table 2). This was in line with Seiberth et al who found no relation between mode of delivery and ROP.²³

Multiple gestation and ROP

In present study, we didn’t find any significant correlation between multiple gestation and occurrence of ROP. Among the study group of 56 babies, 18 were twins. Out of 18 twins, 6 developed ROP (33.33%) compared to 15.79% (6/38) incidence in singleton newborns (Table 2). Thakre in study among 119 neonates found no significant correlation between ROP and twin gestation. ¹²

Association between neonatal factors and ROP

Birth weight and ROP

In our study, 72.73% (8/11) babies born with birth weight less than 1kg developed ROP, while 50% (3/6) babies with birth weight between 1 and 1.5kg developed ROP. The 2.56% (1/39) of babies with birth weight more than 1.5 kg developed ROP (Figure 3) (Table 3). Shetty et al found that all the children with birth weight less than 1500 gm developed ROP.¹⁷ This is similar to our findings that showed decreasing incidence of ROP with increasing gestational age.

Gender and ROP

It was found that 17.86% (5/28) females and 25% (7/28) males developed ROP. There no significant correlation between gender and ROP in our study. Thakare et al also found no correlation between sex and ROP.¹²

Gestational age and ROP

In our study we found that among babies born before 28 weeks, 100% (4/4) developed ROP, while among babies born between 28 weeks-32 weeks of gestation 44.4% (8/18) developed ROP. No case of ROP was reported in neonates born beyond 32 weeks of gestation (Figure 4) (Table 3). A study by Shetty showed an incidence of40% ROP in 30-32 weeks, 10% in 32-34 weeks, and 9.09% in babies above 34 weeks, which was similar to our study.¹⁷

Surfactant and ROP

We found significant correlation between ROP and the use of surfactant; 50% (9/18) of the babies who were treated with surfactant developed ROP (Figure 5) (Table 3). Seibert et al found that surfactant was associated with increasing incidence of ROP.²² In a study by McGinnity et al of the 119 babies who received surfactant treatment incidence of ROP was 45.4% compared to 47.1% in the control group, and no statistical correlation was found.²⁴ Similarly, Rankin et al in their study of 76 babies, with a similar control group, showed that threshold ROP or greater was present in 1.7% of the surfactant group while it was present in 7.8% of the babies in the control group, thus concluding that use of surfactant was not associated with increased risk of ROP. ²⁵

ROP and oxygen supplementation days

We found that 87.5% (7/8) of the babies who required oxygen for more than 28 days developed ROP, 57.14% (4/7) who required oxygen for 10-28 days developed ROP while only 2.44% (1/41) of the babies who required oxygen for less than 10 days developed ROP. Thus, prolonged oxygen supplementation was found to have significant association with incidence of ROP (Table 3). Verma et al found that oxygen therapy and ventilation were significantly associated with the development of ROP.¹⁴ Chaudhari et al found that 64.5% of the babies who received oxygen therapy developed ROP, which was statistically significant.³

Anemia and ROP

In our study 81.82% (9/11) of neonates with anemia developed ROP while only 6.67% (3/45) babies without anemia developed ROP (Table 3). A study by Rekha et al found that apnea and blood transfusions were significantly associated with ROP, which could be due to anemia per se or due to blood transfusions used to treat anemia.⁴ Anemia was found to have statistical correlation with the development of ROP. A study from Vadodara by Patel showed that anemia was a highly significant risk factor for development of ROP.¹

Blood transfusion and ROP

Blood transfusion was also found to be statistically significant risk factor in the development of ROP. The 73.3% (11/15) of the babies who received blood transfusion developed ROP against 2.44% (1/41) among those who did not receive blood transfusion (Table 3). A study on ROP by Thakre from Maharashtra showed significant association between blood transfusion and development ROP.¹²

Phototherapy and ROP

In our study 12 out of 41 who received phototherapy developed ROP (29.27%) whereas none out of 15 babies who didn’t receive phototherapy developed ROP (Table 3). Thus, a statistically significant association was found between phototherapy and development of ROP. This is in contradiction to most of the other studies which reported insignificant association between phototherapy and ROP. Hakeem et al found that there was no significant correlation between the development of ROP and phototherapy.¹⁵ Seiberth et al concluded that the incidence of ROP was decreased in patients who received phototherapy.²³

Sepsis and ROP

Neonatal sepsis was found to have significant correlation with development of ROP. The 43.48% (10/23) babies with history of sepsis developed ROP whereas 6.06% (2/33) without history of sepsis developed ROP (Table 3). In a study from Gujarat by Patel, sepsis was found to be a significant risk factor for ROP.¹ Kumar et al reported higher risk of ROP in children with sepsis.⁵ Sepsis was found to be a culprit in the development of ROP in various other studies as well.^11,12

Limitation

This study is based on a single center and it is the main limitation of the study.

CONCLUSION

Incidence of ROP was found to be 21.4% (12/56) with higher proportion of neonates developing early stages of ROP. The incidence of ROP with a confidence interval of 95% by modified Waid method ranges from 12.5% to 33.9% corresponding to most of the national and international studies, but some studies showed increased incidence of ROP also. Maternal factors including PIH, GDM, Antenatal steroid administration, ante partum hemorrhage, delivery method and multiple gestations were not found to be associated with development of ROP. Neonatal risk factors including lower birth weight, lower gestational age, surfactant therapy, longer duration of oxygen requirement, anemia, blood transfusion, phototherapy and sepsis were found to be major risk factors for development of ROP.

Most of the risk factors associated with ROP arise during the stay in neonatal intensive care unit and some of them are avoidable. The incidence of ROP can be decreased by monitoring the risk factors, timely screening, regular follow up and appropriate interventions.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: The study was approved by the Institutional Ethics Committee

REFERENCES

1. 1. Patel SS, Shendurnikar N. Retinopathy of prematurity in India: incidence, risk factors, outcome and the applicability of current screening criteria. Int J Contemp Pediatr. 2019;6(6):2235-41.
2. 2. Lucey JF, Dangman B. A re-examination of the role of oxygen in retrolental fibroplasia. Pediatrics. 1984;73(1):82-96.
3. 3. Chaudhari S, Patwardhan V, Vaidya U, Kadam S, Kamat A. Retinopathy of prematurity in a tertiary care center--incidence, risk factors and outcome. Indian pediatr. 2009;46(3).
4. 4. Rekha W, Battu RR. Retinopathy of prematurity: incidence and risk factors. Indian pediatrics. 1996;33:999-1004.
5. 5. Kumar N, Kaushik SL, Grover N, Sharma RL. Retinopathy of prematurity: incidence and risk factors: a hospital-based study from Shimla, Himachal Pradesh, India. Int J Res Med Sci. 2017;5(1).
6. 6. International Committee for the Classification of Retinopathy of Prematurity. The international classification of retinopathy of prematurity revisited. Archives of ophthalmology (Chicago, Ill.: 1960). 2005;123(7):991-9.
7. 7. Robinson R, O'Keefe M. Follow-up study on premature infants with and without retinopathy of prematurity. Brit J Ophthalmol. 1993;77(2):91-4.
8. 8. Gilbert C, Rahi J, Eckstein M, O'Sullivan J, Foster A. Retinopathy of prematurity in middle-income countries. The Lancet. 1997;350(9070):12-4.
9. 9. Gilbert C, Fielder A, Gordillo L, Quinn G, Semiglia R, Visintin P, Zin A. 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):e518-25.
10. 10. Gergely K, Gerinec A. Retinopathy of prematurity--epidemics, incidence, prevalence, blindness. Bratislavskelekarskelisty. 2010;111(9):514-7.
11. 11. Braimah IZ, Enweronu-Laryea C, Sackey AH, Kenu E, Agyabeng K, Ofori-Adjei IO et al. Incidence and risk factors of retinopathy of prematurity in Korle-Bu Teaching Hospital: a baseline prospective study. BMJ open. 2020;10(8):e035341.
12. 12. Thakre S, Deshmukh P, Kalyanshetti G, Mishrikotkar J. Incidence, Severity, and Risk Factors of Retinopathy of Prematurity in Central Maharashtra, India. Perinatology. 2017;18(2):51-5.
13. 13. Freitas AM, Mörschbächer R, Thorell MR, Rhoden EL. Incidence and risk factors for retinopathy of prematurity: a retrospective cohort study. Int J Retina Vitreous. 2018;4(1):1-8.
14. 14. Verma A, Nema N, Patel S, Ishrat S, Sidharth M. Retinopathy of prematurity: incidence and risk factors. Int J Ophthalmol Eye Sci. 2016;4:198-201.
15. 15. Hakeem AH, Mohamed GB, Othman MF. Retinopathy of prematurity: a study of prevalence and risk factors. Middle East Afr J Ophthalmol. 2012;19(3):289.
16. 16. Fielder AR, Shaw DE, Robinson J, Ng YK. Natural history of retinopathy of prematurity: a prospective study. Eye. 1992;6(3):233-42.
17. 17. Shetty SP, Shetty J, Amin H, Shenoy RD. The incidence, risk factors and outcome of retinopathy of prematurity at a tertiary care centre in south India. J Dent Med Sci. 2015;14:77-83.
18. 18. Alshaikh B, Salman O, Soliman N, Ells A, Yusuf K. Pre-eclampsia and the risk of retinopathy of prematurity in preterm infants with birth weight <1500 g and/or <31 weeks’ gestation. BMJ open ophthalmol. 2017;1(1):e000049.
19. 19. Yau GS, Lee JW, Tam VT, Liu CC, Wong IY. Risk factors for retinopathy of prematurity in extremely preterm Chinese infants. Med. 2014;93(28).
20. 20. Gebeşçe A, Uslu H, Keleş E, Yildirim A, Gürler B, Yazgan H, Baştürk B, Demirdöven M, Tonbul A. Retinopathy of prematurity: incidence, risk factors, and evaluation of screening criteria. Turk J Med Sci. 2016;46(2):315-20.
21. 21. Kavurt S, Özcan B, Aydemir O, Bas AY, Demirel N. Risk of retinopathy of prematurity in small for gestational age premature infants. Indian pediatr. 2014;51(10):804-6.
22. 22. Shah VA, Yeo CL, Ling YL, Ho LY. Incidence, risk factors of retinopathy of prematurity among very low birth weight infants in Singapore. Ann Acad Med Singapore. 2005;34(2):169-78.
23. 23. Seiberth V, Linderkamp O. Risk factors in retinopathy of prematurity. Ophthalmologica. 2000;214(2):131-5.
24. 24. McGinnity FG, Girschek PK, Morin JD, O'Donoghue P, Dunn MS. Bovine surfactant therapy and retinopathy of prematurity. Ophthalmic Literature. 1995;2(48):110.
25. 25. Rankin SJ, Tubman TR, Halliday HL, Johnston SS. Retinopathy of prematurity in surfactant treated infants. Brit J Ophthalmol. 1992;76(4):202-4.