Risk factors for the development and progression of retinopathy of prematurity in preterm infants in Indonesia

Abstract

BACKGROUND:

Risk factors other than supplemental oxygen might play a role in the development of retinopathy of prematurity (ROP). In Indonesia ROP occurs in infants up to 34 weeks and 2000 g. Risk factors for the development of ROP in Indonesian NICUs have not been evaluated. Our aim was to identify other risk factors than the use of oxygen in the development and progression of ROP in preterm infants in Indonesia.

METHODOLOGY:

Data on 98 preterm infants with ROP and 77 controls were collected from four NICUs and two eye centers in Jakarta, Indonesia, between 2009 and 2014. We used multivariate logistic regression analysis to determine the relationship between infants and environmental variables and the development and progression of ROP. We obtained variables for ROP severity by using Cox regression analysis.

RESULTS:

Factors associated with the development of ROP were birthweight (BWt), intrauterine growth retardation (IUGR), exchange transfusion, duration of oxygen supplementation, minimum saturation monitor setting, and socioeconomic factors. Regarding the progression, gestational age (GA), out-born, duration of supplemental oxygen, minimum saturation monitor setting, and socioeconomic factors were identified as risk factors.

CONCLUSION:

The use and control of supplemental oxygen are the main risk factors for the development and progression of ROP in preterms in Indonesia. Additionally, we confirm that GA, BWt, and IUGR are risk factors. Moreover, we found exchange transfusion to be a risk factor, and we found a lower rate of ROP in infants from a lower socioeconomic background. These risk factors apply to infants with a GA up to 34 weeks and a BWt up to 2000 g.

Keywords

retinopathy of prematurity ROP neonatal risk factors oxygen supplementation

1 Introduction

Over the past few years the number of NICUs and the possibility to treat sick newborn infants has increased significantly in Indonesia. With the introduction of intensive care the survival rate of very preterm infants increased. However, more survivors might also mean an increase in infants with handicaps related to preterm birth and complications arising during treatment. An important complication of very preterm birth is the development of retinopathy of prematurity (ROP). At Harapan Kita Hospital Jakarta, Indonesia we found that 40% of infants born before 28 weeks and 29% of infants born after 28 to 32 weeks developed ROP [1]. In Indonesia the incidence of ROP is higher than in developed countries where ROP is seen mainly in infants born before 28 weeks and hardly in infants with a higher gestational age (GA) [2]. We found that in Indonesia ROP is also seen in late preterm infants [2]. In developed countries the development of ROP is strongly related to GA at birth, birthweight (BWt), and the use of supplemental oxygen [3]. Other factors such as surgical intervention during the neonatal period also influence the risk of developing this disease [3]. To date, no study evaluated which risk factors might play a role in the development of ROP in NICUs in Indonesia. This information is indispensable in the effort to reduce the incidence of ROP in developing countries such as Indonesia.

The aim of this study was to identify whether in Indonesia risk factors other than the use of oxygen contribute to the development of type 2 and the progression to type 1 ROP.

2 Methods

We included all infants admitted to four NICUs in Jakarta, Indonesia, RSAB Harapan Kita, RSIA Budi Kemuliaan, RS Awal Bros Tangerang, and RS Royal Taruma, who were diagnosed with types 1 and 2 ROP and infants with ROP who were referred to the Jakarta Eye Centers Kebon Jeruk and Menteng. The criteria for treating ROP were based on the guidelines for Early Treatment for Retinopathy of Prematurity (ETROP). These guidelines were used to identify infants at high risk of adverse outcomes due to ROP. Type 2 ROP signifies ROP not severe enough to require treatment and type 1 ROP is defined as ROP that does require treatment [4]. Preterm infants without ROP and who were cared for in the same hospitals served as controls. We considered 27 risk factors, including GA, BWt, IUGR, diseases like respiratory distress syndrome, asphyxia, apnea, sepsis, patent ductus arteriosus (PDA), bronchopulmonary dysplasia (BPD), and intraventricular hemorrhage (IVH), the use of oxygen, the setting of the saturation monitor, and diseases of the mother like preeclampsia. The full list of factors we evaluated is given in Table 1. We included the socioeconomic status of the mother because access to medical care is limited in the Indonesian population with a lower socioeconomic background.

Table 1
Potential risk factors for the development and progression of ROP

All 27 variables Potential risk factors for the development of ROP Potential risk factors of the progression of ROP

1 Neonatal characteristics (5):

BWt, GA, IUGR, sex, multiple pregnancy BWt, IUGR, sex BWt, GA, sex

2 Neonatal services (11):

Respiratory distress syndrome, apnea, asphyxia, sepsis, hemodynamic disturbance, BPD, PDA, IVH, PRC transfusion, exchange transfusion, access to hospital Respiratory distress syndrome, asphyxia, hemodynamic disturbance, IVH, exchange transfusion Apnea, access to hospital

3 Oxygen exposure factors (7):

a. Duration of supplemental O₂ (days) (16.5–16.1–<5, no O₂) b. Mean FiO₂ (%) (≥62%, 33 s/d<62%, >21 s/d<33%, 21%) c. Highest FiO₂ (%) Duration of supplemental O₂, mean FiO_2, highest FiO₂, type of supplemental O₂, highest setting SpO₂, lowest setting SpO₂ Duration of supplemental O₂, mean FiO_2, highest FiO₂, type of supplemental O₂, mean setting SpO₂, lowest setting SpO₂ (>40, >21–40, 21)

d. Type of supplemental O₂ (FiO₂ 100%, O₂ blended/FiO₂ 100%, O₂ blended, no O₂) Mean SpO₂ monitor setting

e. (SpO₂≤90%, >90%)

f. The maximum saturation monitor setting (SpO₂ > 93%, ≤93%)

g. The minimum saturation monitor setting – for the development of ROP (SpO₂ < 85%, 85–90%, >90%) – for the progression of ROP (SpO₂≥93.5%, 85 – <93.5%, <85%)

4 Maternal and sociodemographic factors (4):

a. Socioeconomic status (low, middle-upper) Socioeconomic status, mother’s health Socioeconomic status

b. Mother’s health (DM, hypertension, severe preeclampsia, no)

c. Mother’s age

d. Parity

All 27 variables	Potential risk factors for the development of ROP	Potential risk factors of the progression of ROP
1	Neonatal characteristics (5):
	BWt, GA, IUGR, sex, multiple pregnancy	BWt, IUGR, sex	BWt, GA, sex
2	Neonatal services (11):
	Respiratory distress syndrome, apnea, asphyxia, sepsis, hemodynamic disturbance, BPD, PDA, IVH, PRC transfusion, exchange transfusion, access to hospital	Respiratory distress syndrome, asphyxia, hemodynamic disturbance, IVH, exchange transfusion	Apnea, access to hospital
3	Oxygen exposure factors (7):
	a. Duration of supplemental O₂ (days) (16.5–16.1–<5, no O₂) b. Mean FiO₂ (%) (≥62%, 33 s/d<62%, >21 s/d<33%, 21%) c. Highest FiO₂ (%)	Duration of supplemental O₂, mean FiO_2, highest FiO₂, type of supplemental O₂, highest setting SpO₂, lowest setting SpO₂	Duration of supplemental O₂, mean FiO_2, highest FiO₂, type of supplemental O₂, mean setting SpO₂, lowest setting SpO₂ (>40, >21–40, 21)
	d. Type of supplemental O₂ (FiO₂ 100%, O₂ blended/FiO₂ 100%, O₂ blended, no O₂) Mean SpO₂ monitor setting
	e. (SpO₂≤90%, >90%)
	f. The maximum saturation monitor setting (SpO₂ > 93%, ≤93%)
	g. The minimum saturation monitor setting – for the development of ROP (SpO₂ < 85%, 85–90%, >90%) – for the progression of ROP (SpO₂≥93.5%, 85 – <93.5%, <85%)
4	Maternal and sociodemographic factors (4):
	a. Socioeconomic status (low, middle-upper)	Socioeconomic status, mother’s health	Socioeconomic status
	b. Mother’s health (DM, hypertension, severe preeclampsia, no)
	c. Mother’s age
	d. Parity

Abbreviations: ROP – retinopathy of prematurity, GA – gestational age, IUGR – intrauterine growth restriction, BPD – bronchopulmonary dysplasia, PDA – patent ductus arteriosus, PRC – plasma rennin concentration, SpO₂ – oxygen saturation, FiO_2–Inspired oxygen concentration, DM – Diabetes Mellitus.

2.1 Statistical methods

We conducted a clinical study using an epidemiologic approach consisting of two steps. Step I: case control study to evaluate the factors of neonatal management associated with the development of ROP. Step II: using a combination of retrospective and prospective cohort design to evaluate risk factors related to the progression of ROP. In this design exposures are the neonatal risk factors and outcomes are the progression of ROP (type 1 or type 2). The study was approved by the Institutional Review Board of the University of Indonesia School of Public Health. Informed consent was obtained from parents of all infants

Data analysis started with a univariate analysis to determine the patterns and characteristics of the variables, after which the crude odds ratio (COR) was obtained with a bivariate analysis to evaluate the association between potential risk factors and ROP. Subsequently, a multivariate analysis was carried out by logistic regression analysis to obtain the adjusted odds ratio (AOR) and its relation with the development of ROP. This was achieved by compiling all the independent variables with p < 0.25 in the bivariate analysis into multivariate models. Prediction of ROP severity was made with logistic regression, where its variable is obtained from the Cox proportional hazards regression procedure. Cox regression is a procedure that uses a multivariate approach to investigate the effect of several predictor variables on the time it takes for a specified event to occur. This analysis describes the relation between event incidence, as expressed by the hazard function and a set of covariates. We use a hazard ratio to obtain the risk ratio. The hazard ratio resulting from the Cox regression analysis is frequently interpreted as the risk ratio or the relative risk. The difference between the two is that the risk ratio does not take the timing of the event into account but only considers the occurrence of the event at the end of the study period. In this study the outcome of types 1 or 2 ROP was monitored until 44 weeks’ PMA.

We created two different models for the development and progression of ROP: Model 1 is a development and severity model based on variables of oxygen exposure that can be intervened (FiO₂), types of oxygen supplementation, and duration of oxygen supplementation. Model 2 is a development and severity model based on variables that appraise SpO₂. We created these models because the oxygen exposure variables we put into model 1 could change the results of SpO₂ (model 2). Co-linearity existed in the multivariate analysis when the SpO₂ variable (in model 2) was placed in one table with the oxygen exposure variables present in model 1, so we decided to analyze the role of each variable in the two different models.

3 Results

One hundred and seventy-five infants were included in the study: 98 patients suffered ROP and there were 77 controls. Thirty-seven infants had type 1 ROP (38%) and 61 infants had type 2 ROP (62%). The clinical characteristics of the infants are shown in Table 2. Eleven out of thirty-seven (32%) infants with type 1 ROP had a GA of less than 28 weeks compared to 3 of 61 (5%) infants with type 2 ROP (p < 0.001). We found the same trend for BWt, in which case 13 out of 37 (35%) infants with type 1 ROP had a BWt of less than 1000 g, as compared to 11 of 61 (18%) in the type 2 ROP group. There were no significant differences between the infants with types 1 and 2 ROP regarding IUGR and sex (Table 2). We also did not find significant differences between infants with ROP and controls regarding GA, IUGR, and sex. BWt was significantly lower in the group that developed ROP (Table 2).

Table 2
Characteristics of the infants included in the study

Variable Cases (n = 98) Cases (n = 98) Controls (n = 77) P value Case/Control

Type 1 ROP (n = 37) Type 2 ROP (n = 61) P value Type 1/Type 2

Birthweight 0.16 0.02*

Range 529–2 430 676–2 110 529–2 430 600–2 100

Mean (SD) 1 162 (393) 1 247 (298) 1 217 (348) 1 319 (332)

<1000 g 13 11 24 11

1000–1 500 g 20 42 62 44

>1500 g 4 8 12 22

GA 0.001* 0.46

Range 25–34 26–35 24–35

Mean (SD) 30 (3) 31 (2) 31 (3)

<28 wks 12 3 15 8

28–32 wks 15 43 58 44

>32 wks 10 15 25 25

IUGR 0.90 0.07

Prem-SGA 12 19 31 15

Prem-AGA 25 42 67 62

Sex 0.08 0.12

Boys 15 36 51 31

Girls 22 25 47 46

Variable	Cases (n = 98)	Cases (n = 98)	Controls (n = 77)	P value Case/Control
Birthweight			0.16			0.02*
Range	529–2 430	676–2 110		529–2 430	600–2 100
Mean (SD)	1 162 (393)	1 247 (298)		1 217 (348)	1 319 (332)
<1000 g	13	11		24	11
1000–1 500 g	20	42		62	44
>1500 g	4	8		12	22
GA			0.001*			0.46
Range	25–34	26–35			24–35
Mean (SD)	30 (3)	31 (2)			31 (3)
<28 wks	12	3		15	8
28–32 wks	15	43		58	44
>32 wks	10	15		25	25
IUGR			0.90			0.07
Prem-SGA	12	19		31	15
Prem-AGA	25	42		67	62
Sex			0.08			0.12
Boys	15	36		51	31
Girls	22	25		47	46

Abbreviations: ROP – retinopathy of prematurity, GA – gestational age, IUGR – intrauterine growth restriction, prem-SGA – preterm small-for-gestational age, prem-AGA – preterm appropriate-for-gestational age.

The risk factors we found for the development of ROP when using model 1, based on the oxygen variables that can be intervened by a clinician, were IUGR, exchange transfusion, a duration of oxygen exposure of more than 16 days, minimum saturation monitor setting, and socioeconomic factors (Table 3). According to model 2 for identifying risk factors based on the variables that appraise SpO₂ we found BWt, exchange transfusion, minimum saturation monitor setting, and socioeconomic status to be risk factors (Table 3).

Table 3

Risk factors for the development of ROP

Variable	Model 1		Model 2
	P value	AOR (95% CI)	P value	AOR (95% CI)
Birthweight
<1000g			0.07	3.38 (0.90–12.67)
1000 to 1500 g			0.02	3.68 (1.20–11.31)
>1500 g			–	1.00
IUGR
Prem-SGA	0.04	2.38 (1.06–5.38)
Prem-AGA	–	1.00
Sex
Boys	0.09	1.82 (0.92–3.63)
Girls	–	1.00
Respiratory Distress Syndrome
Positive	0.07	0.26 (0.06–1.11)
Negative	–	1.00
Exchange transfusion
Yes	0.01	4.38 (1.42–13.55)	0.004	7.55 (1.90–29.93)
No	–	1.00	–	1.00
Duration of supplemental O₂ administration
>16 days	0.001	5.30 (1.97–14.28)
5 to 16 days	0.09	2.36 (0.87–6.38)
1 to < 5 days	0.42	1.54 (0.54–4.40)
Without O₂ supplementation	–	1.00
Minimum saturation monitor setting
SpO₂ <85%			0.02	0.24 (0.07–0.79)
SpO₂ 85 to 90%			0.88	1.16 (0.17–7.77)
SpO₂ >90%			–	1.00
Socioeconomic status
Lower	0.001	0.30 (0.15–0.63)	0.01	0.34 (0.15–0.76)
Middle – upper	–	1.00	–	1.00

Abbreviations: ROP – retinopathy of prematurity, AOR – adjusted odds ratio, IUGR – intrauterine growth restriction, prem-SGA – preterm small-for-gestational age, prem-AGA – preterm appropriate-for-gestational age, SpO₂ – oxygen saturation.

When using model 1 for the progression of ROP, the risk factors we found included the duration of supplemental oxygen and socioeconomic status (Table 4). Using model 2, the risk factors significantly associated with the progression of ROP were GA, hospital out-versus in-born, and minimum saturation monitor setting (Table 4). Based on our data we calculated that the OR to develop ROP increased by 1.54 when oxygen was given for more than four days, by 2.4 when oxygen was given for five up to 15 days, and by 5.3 when oxygen was given for more than 16 days, as compared to no supplemental oxygen. Infants receiving oxygen for more than seven days had a 4.2 higher risk of developing type 1 ROP. Infants where the minimum setting of the saturation monitor was 93% or higher, had a 2.8 higher risk of developing type 1 ROP, as compared to infants where the saturation was set at 85% or less. The risk to progress to type 1 ROP was higher in out-born compared to in-born infants (Table 4). Out-born infants were more frequently exposed to 100% oxygen and less to blended air and room air. The treatment modalities of infants with severe ROP were not recorded properly in our data. In nearly half of the infants with severe ROP treatment involved laser photocoagulation and/or intravitreal bevacizumab (IVB) injections. Based on information from our ophthalmologists, three infants received IVB therapy, one infant received a combination of IVB therapy and photocoagulation, seven out-born infants received no intervention because at the time of the first examination they were diagnosed with periventricular leukomalacia grades IV and V. In one infant the parents refused intervention. The majority of the remaining infants received laser photocoagulation therapy.

Table 4

Risk factors for the progression of ROP

Variable	Model 1		Model 2
	P value	ARR (95% CI)	P value	ARR (95% CI)
GA
<28 weeks			0.01	2.79 (1.33–5.83)
≥28 weeks			–	1.00
Patient access
Out born infants			0.02	2.32 (1.16–4.65)
In born infants			–	1.00
Socioeconomic status
Lower	0.02	0.36 (0.15–0.86)
Middle – upper	–	1.00
Duration of supplemental O₂ administration
>7 days	0.05	4.16 (0.99–17.41)
1–7 days	0.60	1.54 (0.30–7.95)
Without supplemental O₂	–	1.00
Minimum saturation monitor setting
SpO₂ ≥93,5%			0.18	2.83 (1.19–6.74)
SpO₂ 85 to < 93,5%			0.20	1.70 (0.75–3.83)
SpO₂ < 85%			–	1.00

Abbreviations: ROP – retinopathy of prematurity, ARR – adjusted risk ratio, GA – gestational age, IUGR – intrauterine growth restriction, prem-SGA – preterm small-for-gestational age, prem-AGA – preterm appropriate-for-gestational age, SpO₂ – oxygen saturation.

4 Discussion

In this study we found that BWt, IUGR, exchange transfusion, duration of oxygen supplementation, the setting of the oxygen saturation monitor, and socioeconomic factors were related to the development of type 2 ROP. The progression to severe ROP was related to GA, place of birth, duration of supplemental O₂, the setting of the oxygen saturation monitor, and socioeconomic factors.

A number of studies reported on risk factors for the development of ROP. Recently, Owen and colleagues performed a retrospective cohort analysis of preterm infants referred for ROP screening [3]. They found that GA, BWt, the need for surgery of whatever nature, and maternal magnesium prophylaxis were related to the development of ROP, irrespective of the stage of the disease. Development of severe ROP was related to GA, the need for surgery of whatever nature, and increased probability of death or moderate-severe BPD at 7 days. In 2013 Hellström and colleagues published a comprehensive review on the development and risk factors for ROP [5]. They identified the use of oxygen, neonatal hyperglycemia, low levels of insulin-like growth factor, insufficient nutrition, absence of ω-3 long-chain polyunsaturated fatty acids, and late onset sepsis as risk factors for ROP. Many studies demonstrated the relation between the setting of the oxygen saturation monitor and the development of ROP [6]. Setting the monitor at low levels (85%–89%) was related to a lower incidence of ROP, but an increased death rate [7, 8]. At the same time, setting the monitor between 90%–95% might increase the incidence of ROP [9]. A recent Cochrane Review of five trials including 4965 infants < 28 weeks’ GA targeting the lower levels of SpO₂ compared to the higher levels of SpO₂ found no significant effect on the composite outcome of death or major disability nor on major disability alone, including blindness. There was, however, an increased average risk of mortality of 28 per 1000 infants treated [8, 10].

The oxygen tension, or its substitute the oxygen saturation, is related to the incidence and development of ROP. The SpO₂ monitoring, therefore, is correlated with the incidence and severity of ROP. The SpO₂ monitoring has two monitor settings, the maximum and minimum saturation monitor setting. In this study, we recorded the minimum, mean SpO₂, and maximum oxygen saturation in each sample, both in the cases and controls groups. All three components of SpO₂ are important. When the maximum saturation monitor setting is 95% and the minimum saturation monitor setting 92%, the mean SpO₂ of the infant will be higher than when the minimum saturation monitor setting is 85%. The mean SpO₂ when the saturation monitor is set 95–92% will be at in between these values. In case the range is between 95%–85%, will the resulting saturation be around 90%. In Indonesia, the maximum saturation monitor is almost always set at 95–100%. The SpO₂ in the infant in Indonesia is, therefore, more dependent on the minimum saturation monitor setting, not the maximum saturation monitor setting. This explains why we did not find a correlation between the maximum saturation monitor setting and the incidence of ROP. We found however that the minimum saturation monitor setting was positively correlated with the incidence of ROP, the higher the minimum saturation monitor setting, the higher the incidence and progression of ROP. Our results are in line with other studies on the risk factors for ROP in low income and developing countries. Bas and colleagues [11] conducted a prospective, multicenter study in 69 neonatal intensive care units in Turkey. They also found ROP in infants with a higher BWt and GA. This study showed that red blood cell (RBC) transfusions increased the risks of developing ROP. Transfusions may increase oxygen delivery to the retina because of the lower oxygen affinity of adult hemoglobin in packed red cells. Repeated transfusions may also cause free iron accumulation, which may result in increased production of free hydroxyl radicals, resulting in damage to the retina [12]. We found that the risk of developing ROP was higher in infants who had received two or more transfusions, but this finding was not statistically significant (COR 1.53 (95% CI, 0.70–3.32, P > 0.05). Most likely our finding that exchange transfusion is a risk for developing ROP has the same explanation as RBC transfusions. In 1997 and 2001, respectively, Hesse and colleagues and Dani and colleagues described a correlation between the occurrence of ROP, iron load, and blood transfusions [13, 14]. In 1995, Teoh and colleagues found a correlation between an exchange transfusion and ROP [15]. That we identified exchange transfusion as a risk factor might be related to the more frequent use of exchange transfusions in Indonesia as compared to developed countries. This is due to a relatively high incidence of severe hyperbilirubinemia in Indonesia [16]. More studies from low and middle-income countries found severe ROP in larger preterm infants [17 –21]. The use of 100% oxygen as well as insufficient neonatal care [18 , 22], IVH [21], RDS, sepsis, neonatal jaundice, and oxygen therapy [18] might contribute to the development of severe ROP in larger preterm infants. A number of studies, summarized by Lee, found that sepsis and late onset sepsis in particular might be a risk factor for the development of ROP [23]. We found a small but not significant lower incidence of either proven or suspected sepsis in the control infants (57%) compared to ROP patients (65%). Perhaps our cohorts were too small to detect sepsis as a risk factor. We found that the duration of supplemental O₂ and oxygen saturation (SpO₂≥93.5%) in the minimum saturation monitor setting were the most important factors related to the progression of ROP.

The incidence of ROP is higher in Indonesia as compared to developed countries [2]. ROP is also seen at higher GAs. Despite these differences we found that risk factors for the development of ROP are almost equal to the risk factors found in developed countries. BWt, GA, duration of oxygen supplementation, and the setting of the saturation monitor were significant in the development and progression of ROP. We found that socioeconomic factors were also related to the development of ROP. The risk of infants born to mothers with a lower socioeconomic status to develop ROP is less. Most likely this is related to the finding that oxygen saturation below 89% decreases the incidence of ROP. Infants born to mothers with a lower socioeconomic status are at greater risk of being be born in an environment where supplemental oxygen is not readily available. Lower oxygen levels in plasma might lead to a lower chance of survival but, at the same time, in surviving infants it might lead to a lower risk of developing ROP [7]. If oxygen is readily available, its misuse might induce the development of ROP. Our data stress that ROP is not a disease seen only in very preterm infants. ROP also occurs in infants with a GA up to 34 weeks, which is in accordance to the first descriptions of the disease. In 1952, Silverman and colleagues first described retrolental fibroplasia, which was later named ROP, and found that the incidence of retrolental fibroplasia was highest in infants with a BWt of less than 1250 g, but that the disease also occurred in infants with a BWt up to 2250 g. In this paper the authors described the cohorts of two hospitals in New York. In one hospital the progression to blindness was only seen in infants up to 1500 g, while in the other hospital progression was also seen in infants up to 2000 g [24]. Strict control of the use of oxygen should therefore, not be limited to very preterm infants, but to all preterm infants.

Regarding limitations of this study we mention that we did not evaluate the relation between partial pressure of oxygen (PaO₂) and ROP. In clinical practice it is difficult to measure PaO₂, but it can be estimated from SpO₂ values on the basis that multivariate analyses show that variables of PaO₂ and SpO₂ are collinear. However, there is no good correlation between SpO₂ and PaO₂ above a SpO₂ of 95. A second limitation of our study was that in Step 2 of the research (a combination of retrospective and prospective cohort design) the ROP cases included are the patients who developed ROP (ROP positive) from Step 1 of the case control study, resulting therefore in a larger proportion of type 1 ROP than what it is supposed to be in the population.

5 Conclusion

In Indonesia, similar to developed countries, the duration of supplemental O₂ as well as the minimum saturation monitor setting are important risk factors for the development and progression of ROP. We confirmed that GA, BWt, and IUGR are additional risk factors. Furthermore, we found that exchange transfusion is also a risk factor, while a lower incidence of ROP was found in infants born to mothers with a lower socioeconomic status. These risk factors are also relevant for infants with a GA up to 34 weeks and a BWt up to 2000 g.

Increasing awareness of the above factors is needed in Indonesia and other developing countries to reduce the incidence of ROP to levels found in developed countries.

Competing interest financial disclosure statements

None reported.

Authors’ contributions

As principle investigator Dr. Siswanto had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Siswanto, Ronoatmodjo

Acquisition of data: Adisasmita

Analysis and interpretation of data: Siswanto, Sitorus (ophthalmology section)

Drafting of the manuscript: Siswanto, Ronoatmodjo, and Soemantri (first draft), Sauer (second draft and discussion)

Critical revision of the manuscript for important intellectual content: Siswanto, Sauer

Statistical analysis: Adisasmita, Siswanto, and Soemantri

Study supervision: Siswanto, Sauer

Obtained funding

None.

Footnotes

Acknowledgments

This work was carried out and supported by the Harapan Kita Children’s and Women’s Hospital, the Eye Centers Kebon Jeruk and Menteng, and Budi Kemuliaan Hospital, all situated in Jakarta. Indonesia. We would like to thank all doctors, especially Nani H. Widodo MD and Florence Manurung MD, who conducted the ophthalmologic examination in the preterm infants whom we studied in this research. We thank Dr. T. van Wulfften Palthe for correcting the English language.

References

Siswanto

, Widodo

, Sauer

PJJ

. Eleven years of Retinopathy of Prematurity in one NICU in Jakarta, Indonesia. Arch Dis Child. 2018;0:1–3.

Siswanto

, Sauer

PJJ

. Retinopathy of prematurity in Indonesia: Incidence and risk factors. JNPM. 2017;10:85–90.

Owen

, Morrison

, Hoffman

, Yoder

, De Angelis

. Retinopathy of prematurity: a comprehensive risk analysis for prevention and prediction of disease. PLoS One. 2017;12:e0171467.

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;121:1684–9.

Hellström

, Smith

, Dammann

. Retinopathy of prematurity. Lancet. 2013;382:1445–57.

SUPPORT Study Group of the Eunice Kennedy Shriver NICHD Neonatal Research Network, Carlo

, Finer

, Walsh

, Rich

, Gantz

, et al. Target ranges of oxygen saturation in extremely preterm infants. N Engl J Med. 2010;362:1959–69.

Saugstad

, Aune

. Optimal oxygenation of extremely low birth weight infants: a meta-analysis and systematic review of the oxygen saturation target studies. Neonatology. 2014;105:55–63.

Askie

, et al. Effects of targeting lower versus higher arterial oxygen saturations on death or disability in preterm infants. Cochrane Database Syst Rev. 2017;4:CD011190.

Sola

, Golombek

, Montes Bueno

, Lemus-Varela

, Zuluaga

, Domínguez

, et al. Safe oxygen saturation targeting and monitoring in preterm infants: can we avoid hypoxia and hyperoxia? Acta Paediatr. 2014;103:1009–18.

10.

Cayabyab

, Arora

, Wertheimer

, Durand

, Ramanathan

. Graded oxygen saturation targets and retinopathy of prematurity in extremely preterm infants. Pediatr Res. 2016;80:401–6.

11.

Bas

, Demirel

, Koc

, Isik

, Hirfanoglu

, Tunc

. TR-ROP Study GrouIncidence, risk factors and severity of retinopathy of prematurity in Turkey (TR-ROP study): a prospective, multicentre study in 69 neonatal intensive care units. Group. Incidence, risk factors and severity of retinopathy of prematurity in Turkey (TR-ROP study): a prospective, multicentre study in 69 neonatal intensive care units. Br J Ophthalmol. 2018;0:1–6.

12.

Wardle

, Drury

, Garr

, Weindling

. Effect of blood transfusion on lipid peroxidation in preterm infants. Arch Dis Child Fetal Neonatal Ed. 2002;86:46F–8.

13.

Hesse

, Eberl

, Schlaud

, Poets

. Blood transfusion. Iron load and retinopathy of prematurity. Eur J Pediatr. 1997;156:465–70.

14.

Dani

, Reali

, Bertini

, Martelli

, Pezzati

, Rubaltelli

. The role of blood transfusions and iron intake on retinopathy of prematurity. Early Hum Dev. 2001;62:57–63.

15.

Teoh

, Boo

, Ong

, Nyein

, Lye

, Au

. Duration of oxygen therapy and exchange transfusion as risk factors associated with retinopathy of prematurity in very low birthweight infants. Eye. 1995;9:733–37.

16.

Greco

, Arnolda

, Boo

, Iskander

, Okolo

, Rohsiswatmo

, et al. Neonatal jaundice in low-and middle-income countries: Lessons and future directions from the 2015 Don Ostrow Trieste Yellow Retreat. Neonatology. 2016;110:172–80.

17.

, Zhou

, Zhang

, Ji

, Zhang

, Zhu

, et al. Screening for retinopathy of prematurity in China: a neonatal units-based prospective study. Invest Ophthalmol Vis Sci. 2013;54:8229–36.

18.

Hungi

, Vinekar

, Datti

, Kariyappa

, Braganza

, Chinnaiah

, et al. Retinopathy of prematurity in a rural neonatal intensive care unit in South India—a prospective study. Indian J Pediatr. 2012;79:911–5.

19.

Shan

, Ni

, Xue

, Yu

, Huang

. Type 1 retinopathy of prematurity and its laser treatment of large preterm infants in East China. PLoS One. 2015;10:e0144313.

20.

Shah

, Narendran

, Kalpana

. Aggressive posterior retinopathy of prematurity in large preterm babies in South India. Arch Dis Child Fetal Neonatal Ed. 2012;97:F371–5.

21.

Chen

, Lien

, Lai

, Chao

, Chen

, Hwang

, et al. Retinopathy of prematurity in neonatal patients with birth weight greater than 1500g in Taiwan. Biomed J. 2013;36:84–9.

22.

Gunay

, Celik

, Tuten

, Karatekin

, Bardak

, Ovali

. Characteristics of severe retinopathy of prematurity in infants with birth weight above 1500 grams at a referral center in Turkey. PLoS ONE. 2016;11:e0161692.

23.

Lee

, Dammann

. Perinatal infection, inflammation, and retinopathy of prematurity. Semin Fetal Neonatal Med. 2012;17(1):26–9.

24.

Silverman

, Blodi

, Locke

, Day

, Reese

. Incidence of retrolental fibroplasia in New York nursery. AMA Arch Ophthalmol. 1952;48:698–711.