Retinopathy of Prematurity Incidence and Risk Factors in a Tertiary Hospital in Riyadh, Saudi Arabia.

PURPOSE: The purpose of the study was to determine the incidence of retinopathy of prematurity (ROP) in preterm infants in the neonatal intensive care unit (NICU), to identify the risk factors that predispose to ROP, and to assess the outcome of these infants.
MATERIALS AND METHODS: This was a retrospective cohort study on premature infants with a birth weight (BW) of <1501 g or gestational age (GA) of <32 weeks. These infants were admitted to the NICU at a tertiary hospital in Riyadh, Saudi Arabia, from January 2010 to December 2014.
RESULTS: Five hundred and ninety-three infants were included; data were available for 581 infants. Of them, 224 infants (38.6%) had ROP. Of these, 22 infants (10.4%) had stage 3 ROP. The mean BW of infants with ROP was 938.4 ± 257.9 g, and the mean GA at birth was 27 ± 2.4 weeks. A significant relationship with a P < 0.05 was found between the occurrence of ROP and small GA at birth, low BW, low APGAR score at 1 min, and long duration of receiving oxygen (O2) therapy. Based on charts review, none of the infants had blindness.
CONCLUSION: The incidence of ROP in our study falls in the range of incidence in developing countries. The low BW and small GA were the most significant risk factors. Furthermore, it is also recommended to control the amount and duration of O2 therapy to as little as needed. Copyright:

Introduction

Retinopathy of prematurity (ROP) can be considered as one of the most common causes of blindness in children.[1] It is defined as “a potentially blinding eye disorder that primarily affects premature infants weighing about 1250 g or less that are born before 31 weeks of gestation.”[2] ROP was diagnosed for the first time in 1942 by Terry, so it was previously called “Terry syndrome.”.[3] The development of ROP is due to the formation of immature vasculature in the retina of premature infants which might lead to retinal detachment and blindness. There are two phases of how these immature vessels formed. First, a hyperoxic phase causes constriction of the retinal vessels and ischemia. Second, this ischemia might lead to angiogenic factor production by the mesenchymal spindle cells and the formation of immature new vessels subsequently.[45] The International Classification of ROP was formed in 1984. It classifies the ROP according to the location of the disease into zones of the retina (1, 2, and 3), the extent of the disease based on the clock hours (1–12), and the severity of the disease into different stages (0–5).[6]

Few studies have been conducted in the Kingdom of Saudi Arabia; a retrospective study was conducted by Binkhathlan et al. in 2008, in which 174 infants were included from the neonatal intensive care unit (NICU) in Riyadh. The incidence of ROP in their study was 56%; 15% of those patients had Stage 3 ROP. The mean gestational age (GA) was 30 weeks for the ROP-positive group.[7] In another study carried out in 2003 at a University Hospital in Riyadh, it included 195 consecutive preterm infants in the NICU during the period from 1995 to 1998. The study found that the overall incidence of ROP was 37.4%. A percentage of 26% of these children who were diagnosed with ROP reached threshold ROP and needed laser treatment or cryotherapy that induced regression in all of the patients.[8]

Risk factors for ROP have been the main concern in many previous studies. Most of the suggested risk factors include preterm delivery, low birth weight (BW), and excessive oxygen (O2) supplementation to the baby that might affect the immature retina.[9] Other risk factors might occur in the postnatal period of the baby's life. These include neonatal bacteremia (sepsis), anemia, hyperglycemia, and chronic lung disease.[1011] Shastry et al. suggested that genetic predisposition is an important risk factor of ROP.[12] Early diagnosis and screening measures are very important aspects of treating ROP since early intervention can improve visual outcomes. Detailed fundus examination for all babies who have a recognized risk factor for ROP is the mainstay of screening. Failure of diagnosis and proper management can result in loss of vision, amblyopia, strabismus, and retinal detachment.[13]

The purpose of this study was to assess the incidence of ROP, its risk factors, complications, and the effects of comorbidity on the outcome of the treatment. The findings of this study will provide a good update to local studies and help in planning better approaches to improve the management of the ROP.

Materials and Methods

This was a hospital-based, retrospective cohort study conducted in the neonatology department at a tertiary hospital in King Abdulaziz Medical City, Riyadh, Saudi Arabia. It is a 1000-bed specialized health-care institution that covers a wide range of secondary and tertiary care specialties. Neonatology division is one of the biggest neonatology divisions in the whole kingdom, with the number of deliveries reaching 9,000 live births per year. It encompasses a Level III B NICU, with 40-ventilated NICU bed capacities, aside from the 70 beds in the special care nurseries.

Saudi and non-Saudi male and female babies who were admitted to the NICU between January 1, 2010, and December 30, 2014, were studied. We restricted our sample for patients who have had BW <1500 g or born at <32 weeks GA. Furthermore, all infants who received >49 days of continuous O2 therapy were included in the study. Infants with major congenital anomalies and those who died before the discharge from NICU were excluded from the study. Data that were collected for the babies included the date of birth, sex, GA at birth, and BW. We also recorded all relevant information including concomitant conditions such as interventricular hemorrhage, sepsis, necrotizing enterocolitis (NEC), duration of O2 therapy and mechanical ventilation, and pre- and postnatal medications. Diagnosis, stage, zone, and mode of treatment of ROP were also documented.

All data variables were managed and analyzed by IBM SPSS software. Continuous variables such as age and BW were expressed as mean ± standard deviation. Categorical variables such as gender and the comorbidities were presented as frequencies and percentages. To evaluate the risk factors, the Chi-square test or Fisher's exact test has been applied. P < 0.05 was considered to show a significant difference.

This study was performed following the ethical standards of the 1964 Helsinki Declaration, and ethical approval was obtained from the Institutional Research Board at King Abdulaziz Medical City, Riyadh, Saudi Arabia.

Results

A total of 3146 infants were admitted to the NICU during the study period. A number of 593 infants met our inclusion criteria. Among which, 272 infants were males (45.9%). Of the 593 infants, there were 197 (33.2%) normal spontaneous vaginal delivery, 43 (7.3%) assisted delivery, 345 (58.2%) were Cesarean Section, and 8 (1.3%) were not documented. The mean BW was 1118 ± 276.6 g (range: 430–2280 g) and the mean GA at birth was 28.8 ± 2.8 weeks (range: 22–39 weeks). Of 585 infants, there were 367 (61.9%) singletons, while 218 (38.1%) were the results of multiple pregnancies.

ROP was diagnosed in 224 (38.6%) of the 581 babies who were evaluated for ROP, of these, 212 were staged for ROP and 22 (10.4%) had Stage 3, i.e., severe disease [Figure 1]. The mean GA of infants with ROP was 27 ± 2.4 weeks, and the mean BW was 938.4 ± 257.9 g (median = 907.5, range = 430–2280). Both eyes were affected by ROP in 131 infants (60%), isolated left eyes were affected in 42 (19%) infants, and isolated right eyes were affected in 45 (21%) infants. Three infants (1.5%) developed threshold ROP requiring treatment, i.e., Stage 3 in Zone 1 or 2 with plus disease and at least 5 contiguous or 8 cumulative clock hours of involvement. The comparison of the BW and GA of infants with and without ROP is shown in Tables 1 and 2, respectively. It was found that infants with ROP were more likely to have BWs <1000 g and GA <28 weeks as compared to infants without ROP (P < 0.001).

Figure 1

Distribution of infants’ admissions to NICU. *No ROP includes 12 cases that had missing data for ROP. NICU; Neonatal intensive care unit, ROP; Retinopathy of prematurity

Table 1

Birth weight distribution of the study population.

Birth weight (g)	ROP, n (%)	No ROP, n* (%)	Total, n (%)
≤750	55 (25)	11 (3)	66 (11)
751-1000	89 (40)	55 (15)	144 (25)
1001-1250	53 (24)	100 (28)	153 (26)
1251-1500	25 (11)	169 (48)	194 (33)
>1500	2 (1)	21 (6)	23 (4)
Total	224 (100)	356 (100)	580 (100)

*One baby from no ROP group had missing birth weight. P<0.001 for comparison of birthweight distribution between the two groups. ROP: Retinopathy of prematurity

Table 2

Gestational age distribution of the study population

Gestational age (weeks)	ROP, n (%)	No ROP, n (%)	Total, n (%)
<28	137 (61)	64 (18)	201 (35)
28-29	56 (28)	104 (29)	160 (28)
30-31	18 (8)	103 (29)	121 (21)
≥32	12 (5)	86 (24)	98 (17)
Total	223* (100)	357 (100)	580 (100)

*One baby from ROP group had missing gestational age. P<0.001 for comparison of gestational age distribution between the two groups. ROP: Retinopathy of prematurity

The significant risk factors for ROP (P < 0.05) included GA at birth (P < 0.001), BW (P < 0.001), APGAR score at 1 min after birth (P = 0.003), duration of receiving O2 therapy (P < 0.0001), and duration of mechanical ventilation (P < 0.0001). The insignificant risk factors for ROP (P ≥ 0.05) included the age of the mother at the time of delivery (P = 0.704), how many babies the mother was pregnant with (P = 0.051), and the APGAR scores at 5 min and 10 min (P = 0.190) and (P = 0.280), respectively.

All of the risk factors were included in multiple backward logistic regression analysis. BW < 1500 g at birth was found to be the most significant risk factor for ROP, (odds ratio [OR] = 11.6, 95% confidence interval [CI] from 1.07 to 126.02, P = 0.04), followed by GA at birth < 32 weeks (OR = 3.35, 95% CI from 1.22 to 9.16, P = 0.02). Other risk factors are shown in Table 3 with OR and P value.

Table 3

Risk factors odds ratio and its respective confidence intervals

Risk factor	OR	95% CI of OR	P
Antenatal steroid	0.96	0.52-1.76	0.892
Postnatal steroid	0.46	0.27-0.81	0.007*
Surfactant therapy	0.75	0.44-1.29	0.294
Septicemia	0.64	0.32-1.26	0.196
O2 therapy	0.39	0.15-0.77	0.001*
NEC	1.51	0.61-3.76	0.376
PDA	1.22	0.70-2.13	0.486
IVH	0.55	0.32-0.93	0.026*
Packed RBC	0.41	0.20-0.86	0.019*
Jaundice	0.91	0.52-1.61	0.746
Hyperglycemia	1.22	0.26-5.83	0.803

*Significant P value. OR: Odds ratio, CI: Confidence interval, O2: Oxygen, NEC: Necrotizing enterocolitis, PDA: Patent ductus arteriosus, IVH: Intraventricular hemorrhage, RBC: Red blood cell

Among those 224 infants with ROP, three infants had a laser treatment and 12 had anti-vascular endothelial growth factor (VEGF). The most commonly used anti-VEGF treatment for ROP in our center was ranibizumab (Lucentis). Based on the chart review after at least 3 months of discharge, no one of those infants had blindness.

During the review of charts, five patients have been found to have ROP with a GA of more than 32 weeks. One of them had 35, two of them had 34, and two had 33 weeks. In addition, a number of two patients have been found to have ROP with a BW of more than 1500 g.

Discussion

In our tertiary hospital, we follow the Ministry of Health Guidelines for ROP screening. These guidelines identify premature infants with GA at birth of <32 weeks and BW of <1500 g as potentials at risk of ROP.[14] Moreover, those patients who had other comorbidities, for example, O2 therapy, surfactant use, septicemia, and NEC, were of special attention. A full dilated fundus examination with the use of lid speculum and scleral depressor has been done for every premature infant at the postmenstrual age of 31 weeks or 4–6 weeks chronological (postnatal) age whichever is later. Follow-up examinations were recommended by the examining ophthalmologist on the basis of retinal findings classified according to the International Classification.[6]

The universal reported incidence of ROP among infants with risk ranges between 29 and 68%.[78151617] In this study, 37.8% of the patients were diagnosed with ROP, and this is similar to what has been reported by Al-Amro et al. in 2003 but much lesser than the percentage found in Binkhathlan et al. study in 2008[78] due to the larger sample size, longer review period (2010–2014), and the concordance with one of the local studies; we believe that our results are targeting the correct incidence in the Kingdom of Saudi Arabia.

Among the study population, the difference in the proportion of males and females who had ROP in the study period was statistically insignificant. This is concordant with a study that has been done in Iran by Feghhi et al. in 2012, with the results showing insignificant male/female ratio.[18] Furthermore, the number of babies during the pregnancy either singleton or multiple has shown to be insignificant in this study, which is also concordant with the previous Iranian study.

The occurrence of some risk factors for ROP has been found to be significant among most of the studies including this study. Those factors included small GA at birth, low BW, low APGAR score at 1 min after birth, and long duration of receiving O2 therapy.[789101115161718] As our NICU follows a conservative O2 saturation policy, i.e., 88%–92%, the ROP incidence was found to be related to the duration of O2 therapy more than the amount.

During the review of charts, five patients have been found to have ROP with GA more than 32 weeks. One of them had 35, two of them had 34, and two had 33 weeks. In addition, a number of two patients have been found to have ROP with BW more than 1500 g. A similar observation has been noted by Binkhathlan et al. in 2008.[7] These findings indicate that the current screening criteria in Saudi Arabia might need some modifications to involve such cases by widening the inclusion criteria to include infants with GA more than 32 weeks and those born with BW more than 1500 g.

Since the advances in treating ROP with laser and more recently with anti-VEGF for threshold ROP, a good prognosis is expected. Moreover, while the blindness caused by ROP is a serious issue and at least 50,000 infants are blind from ROP globally,[19] none of the patients with ROP ended up with blindness in this study. We believe that careful screening and extensive proper management are good measures for those types of patients.

This study had some avoidable limitations. In addition, being a retrospective study with all the known limitations of such study design, all the patients were from only one hospital and with a relatively short follow-up duration of time.

Conclusion

ROP is one of the major diseases affecting premature infants. The incidence of ROP in our study is comparable to other local studies and falls in the range of disease incidence in developing countries. While the low BW and small GA were the most significant risk factors, the screening criteria of ROP should be broadened at least to include those infants with GAs between 32 and 35 weeks. Furthermore, it is also recommended to control the amount and duration of O2 therapy to as little as needed. Although blindness is well-known consequence of ROP, it has not been encountered in our study population.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.