|Home | About | Journals | Submit | Contact Us | Français|
The purpose of this study is to determine the incidence of retinopathy of prematurity (ROP) and the maternal/neonatal risk factors at a tertiary care hospital in Oman, compared to other countries.
A retrospective analysis of premature neonates born with gestational age (GA) 24–32 weeks at Sultan Qaboos University Hospital, Oman, from January 2007 to December 2010. Maternal and neonatal in-hospital course was retrieved. The incidence of ROP was reported. Risk factors analyses were performed using univariate and multivariate statistics.
A total of 171 neonates (57% males, 43% females) were included for analysis. The incidence of ROP (any stage) was 69/171 (40.4%). Infants with ROP had significantly lower GA (27.7±2 weeks) compared to non-ROP group (30.2±1.7 weeks), P < 0.001),P < 0.001) and significantly lower birth weight (BW) (948 ± 242 g in ROP group vs. 1348 ± 283 g in non-ROP group;P < 0.001). Other significant risk factors associated with ROP were: small for GA, respiratory distress syndrome, requirement for ventilation, duration of ventilation or oxygen therapy, bronchopulmonary dysplasia, hyperglycemia, late onset sepsis (clinical or proven), necrotizing enterocolitis, patent ductus arteriosus, seizures, and number of blood transfusions. There was no significant difference in maternal characteristics between the ROP and non-ROP groups except that mothers of infants with ROP were found to be significantly younger. Logistic regression analysis revealed early GA, low BW, duration of Oxygen therapy, and late-onset clinical or proven sepsis as independent risk factors.
ROP is still commonly encountered in neonatal practice in Oman and other countries. Early GA, low BW, and prolonged oxygen therapy continue to be the main risk factors associated with the occurrence of ROP in our setting. In addition, an important preventable risk factor identified in our cohort includes clinical or proven late-onset sepsis.
Retinopathy of prematurity (ROP) is a vasoproliferative disorder of the retina occurring in premature neonates due to the presence of immature retinal vasculature and is the leading cause of blindness. It is a multifactorial disease process with the most significant independent risk factors for its development include prematurity, low birth weight (BW), and prolonged oxygen therapy.[2,3,4,5,6,7,8,9] Hyperoxia in the neonatal intensive care is a major factor implicated in the causation of ROP. Other risk factors and comorbidities identified include sepsis, patent ductus arteriosus (PDA), intraventricular hemorrhage (IVH), hyperglycemia, poor weight gain, total parenteral nutrition, and blood transfusion.[4,5,7,8,9]
Despite the advances in antenatal and neonatal therapeutic interventions, screening, and follow-up, ROP remains a potentially vision-threatening retinopathy requiring vigilant surveillance and timely intervention to prevent progression to adverse visual impairment or blindness. The prevalence of ROP varies in different populations, races, and neonatal units. In Oman, an incidence of ROP in preterm neonates was documented as 34% in 1994 and 25% in 2003,[2,3] however, there is no follow-up study to document the change in incidence with the current Neonatal Intensive Care Unit (NICU) practices. The increasing fertility rate and improving survival of more extreme preterm neonates has resulted in increasing number of prematurity complications encountered in the neonatal population.
It is important to establish the risk factors for ROP occurrence in the local population, as those can be modified to reduce the incidence of this vision-threatening condition. The objective of the current study is to assess the incidence of ROP in preterm neonates admitted to one of the main tertiary hospitals in Oman and compare it to local and international incidence rates. In addition, we aimed to assess its risk factors by analyzing maternal and neonatal characteristics and to compare the results with other reports from the Middle East and other countries worldwide.
This is a retrospective cohort study over a 4-year period (January 2007 to December 2010) at NICU, Sultan Qaboos University Hospital (SQUH), Muscat, Oman. The inclusion criteria were all neonates who were screened for ROP and survived up to hospital discharge. Patients who died during the hospital course, or had incomplete clinical data for assessment of risk factors, or missed screening for ROP, were excluded from the study.
Screening for ROP was done by either one of the two pediatric ophthalmologists in SQUH. The ophthalmological examinations were first performed at 30–31 weeks postmenstrual age in neonates born at <28 weeks gestational age (GA) and at 4–5 weeks postnatal age in neonates born at 28–32 GA, and were repeated weekly or biweekly, according to the recommendations by the American Academy of Pediatrics (AAP) Section on Ophthalmology; American Academy of Ophthalmology (AAO); American Association for Pediatric Ophthalmology and Strabismus (AAPOS), American Association of Certified Orthoptists, until vascularization of the retina reached ora serrata (the most peripheral part of the retina), or until full remission of ROP after treatment.[11,12] The initial recommendation of 2006 was to screen all neonates born below 32 weeks of gestation, followed by 2013 guidelines that recommends to screen infants with GA of 30 weeks or less and selected infants with GA of >30 weeks with an unstable clinical course, including those requiring cardiorespiratory support and who are at high risk for ROP. Although at the time of our study period the unit followed 2006 recommendations that advised to screen all neonates below 32 weeks of gestation, our study patients selection is also applicable to the recent 2013 recommendations as all neonates more than 30 weeks of gestation in our cohort were admitted to the unit with unstable course (being a tertiary care referral center). During the examination procedure, eyes were dilated with a combination of cyclopentolate 0.2% and phenylephrine 1% eye drops, one drop every 10 min for 2 times. The ophthalmological examination included indirect ophthalmoscopy with a 28D lens that was performed for each neonate using speculum and scleral depression. The findings of retinal examination were documented. RetCam 3 fundus imaging was done when indicated.
The demographic and clinical data with regards to maternal and neonatal risk factors were obtained from the hospital electronic patients' records. The maternal variables that were followed included age, gravidity, multiple pregnancy, antepartum hemorrhage, gestational diabetes, gestational hypertension, antenatal steroids, preeclampsia, and premature rupture of membrane. The neonatal factors studied were GA based on the last menstrual period, BW and its centile, sex, respiratory distress syndrome (RDS), hypotension, hyperglycemia, total duration of oxygen use, IVH, significant PDA confirmed by echocardiogram, and bronchopulmonary dysplasia (BPD). Hypotension was defined as mean arterial pressure reading < number of weeks of GA and/or clinical signs of hypoperfusion. Hyperglycemia was defined as bedside glucose >8 mmol/L. Other variables included were late onset sepsis (either proven by microorganisms isolated in blood culture or based on clinical suspicion that required at least 5 days of treatment), number of blood transfusions in the first 4 weeks of life and percentage postnatal weight gain by 4 weeks of age.
The ethical approval for this study was obtained from the Research and Ethics Committee at Sultan Qaboos University.
Data were analyzed using SPSS (Statistical Package for Social Sciences) software, version 20 (IBM, Chicago, Illinois, USA). The incidence of ROP was calculated in the whole cohort and then separately for different groups according to the GA and BW. Mean (±standard deviation [SD]) or median (with interquartile range) were used to present the numerical data, wherever appropriate. Risk factors for ROP were estimated using univariate and multivariate analyses. For univariate analysis, comparison between ROP and non-ROP groups was done using Chi-square test for categorical variables, and Student's t-test or Wilcoxon-Mann-Whitney for numerical variables according to their normal distribution. A two-tailed level (P value) was set at 0.05 level for significance. A logistic regression model was performed to all studied risk factors and the adjusted odds ratio (OR) (95% confidence interval [CI]) was obtained.
Among a total of 1,857 neonates who were admitted to NICU during the 4-year study period (January 2007 to December 2010), 233 were evaluated for eligibility to be included in this study (GA ≤30 weeks, or >30 weeks with unstable course). Sixty-two patients were excluded from the study (31 expired before discharge, and 31 were excluded because they had incomplete clinical data, or lost to follow-up). The remaining 171 neonates were included for analysis: 97 (57%) males and 74 (43%) females, with a mean (±SD) BW of 1200 (±330) grams, and GA of 30 (±2) weeks.
Out of the 171 preterm neonates included in the study, 69 developed ROP (of any stage) in one or both eyes, giving the incidence of 40.4% during the 4-year study duration. The overall incidence of ROP and specific incidence according to different GA and BW groups are shown in Figures Figures11 and and2.2. Higher incidence of ROP was found with neonates with lower GA and lower BW. None of the neonates in the study presented with ROP Stage 4 or 5.
Comparison in demographic and clinical characteristics between ROP and non-ROP groups is shown in Table 1. Univariate analysis demonstrated a significantly lower GA in neonates with ROP ROP than those without ROP (27.7 ± 2 weeks versus 30.2 ± 1.7 weeks, respectively; P < 0.001) and lower BW (948 ± 242 g in ROP group versus 1348 ± 283 g in non-ROP group; P < 0.001). In addition, neonates with ROP were found to have significantly younger maternal age (26.5 ± 4.0 years versus 29.0 ± 5.4 years, P < 0.001) and were more frequently small for GA compared to those without ROP (25% versus 11%; P < 0.001).
The course and comorbidities during intensive care of the neonates with ROP and those without ROP are shown in Table 2. Univariate analysis indicated that the significant risk factors associated with ROP included: RDS, surfactant therapy, requirement for ventilation, duration of invasive ventilation or nasal continuous positive airway pressure (CPAP), duration of oxygen therapy, occurrence of BPD, hyperglycemia, presence of late onset sepsis (clinical or proven), Gram-negative sepsis, necrotizing enterocolitis (NEC), PDA, occurrence of seizures, and number of blood transfusion (P < 0.05).
Analysis of risk factors using multivariate logistic regression revealed that the significant adjusted risk factors were: Early GA (OR: 2.106; 95% CI: 1.029–4.313; P = 0.042), low BW (OR: 1.005; 95% CI: 1.001–1.009; P = 0.014), duration of invasive ventilation (OR: 1.192; 95% CI: 1.020–1.393; P = 0.027), duration of CPAP (OR: 1.206; 95% CI: 1.041–1.398; P = 0.013), duration of oxygen therapy (OR: 1.113; 95% CI: 1.023–1.212; P = 0.013), and presence of late-onset clinical or proven sepsis (OR: 7.472; 95% CI: 1.212–46.047; P = 0.030) [Table 3].
The present study revealed an overall incidence of ROP as 69/171 (40.4%) over the 4-year duration. This represents an increasing trend compared with two previous studies in SQUH that reported the incidence of 34% and 25.4% in 1994 and 2003, respectively.[2,3] This could be attributed to the younger age of premature neonates delivered at our hospital, which is reflected by the inclusion of 24–25 weeks preterm babies (BW 500–750 g) that were not present in the previous studies. In addition, we had to exclude the patients who expired during the neonatal course, and those who had incomplete clinical data or lost to follow-up.
Comparing ROP incidence in Oman versus other countries in the Middle East, the reports from these countries varied according to inclusion criteria and studied population. Studies of premature newborns from the United Arab Emirates and Saudi Arabia reported ROP incidence of 11% and 28%, respectively, which are lower than the incidence in our study.[13,14] However, patients included were of higher GA. Hadi and Hamdy reported ROP incidence of 34% from Egypt, which is close to the reported incidence from Iran as 32% both of which are slightly lower than our reported incidence. Other international studies have reported a wide range of ROP incidence, starting from as low as 29.2% in Singapore to as high as 47% in the USA, as illustrated in Table 4. The variation in ROP incidence may be attributed to the difference in neonatal practice, studied subjects, as well as genetic and racial background. Although during the era of our study, the guideline available from AAP, AAO, AAPOS was only the one published in 2006 and recommended screening in all neonates ≤32 weeks of gestation, our cohort matched 2013 recommendations as it included neonates >30 weeks GA and those >30 weeks GA but admitted with unstable course (being a tertiary care referral center).
The incidence of ROP according to GA was the maximum in neonates born at 24 weeks of gestation (100%) and gradually decreased to a range between 60% and 87% in those born at 25–28 weeks of gestation. The incidence at older GA groups gradually decreased in the subsequent older gestational, to as low as 4% in those 32 weeks of gestation. Lower GA as a risk factor was proved significant in our study, even after regression analysis module (P = 0.042, OR 2.11, 95% CI = 1.03–4.31; for each 1 week younger as GA). The increasing incidence of ROP in premature neonates is possibly due to interruption of normal retinal angiogenesis. Most studies have consistently shown the prematurity to be the strongest predictive factor of ROP suggesting that factors related to growth and development are critical to ROP disease process.[2,3,4,5,6,7,8,9]
Low BW was found to be significantly correlated with the incidence of ROP in our study, after multivariate regression module (P = 0.014, OR 1.65, 95% CI = 1.11–2.45; for each 100 g less as BW), which is in agreement with many other studies.[2,3,4,5,6,7,8,9] The third important risk factor in our study is prolonged oxygen exposure by invasive or noninvasive ventilation, and by nasal cannula, that was proportionately increasing the risk of ROP occurrence. The premature neonate is developmentally unprepared to the oxygen-rich extrauterine life because of an impaired antioxidant defense system.[16,17] Oxidative stress, a result of oxygen free radicals generation after exposure to high oxygen concentration affects different organs simultaneously and has been implicated in the development of many neonatal diseases such as ROP, BPD, NEC, IVH, and periventricular leukomalacia.[18,19]
Late onset neonatal sepsis was also found as a significant risk factor for ROP in our study. This is in agreement with other studies.[4,5,8] Sepsis is well recognized to be present with high incidence in preterm neonates, due to immature innate immunity resulting in a profound defect in the recognition and the clearance of microorganisms.[20,21] A study by Dammann et al. demonstrated that sepsis-associated increased risk for any stage of ROP might be due to circulating inflammatory mediators. Other investigators studied the association of perinatal inflammation as evaluated by cytokine levels and ROP, with the conclusion that elevated blood concentrations of multiple pro-inflammatory cytokines during the first few weeks of postnatal life in newborns who develop ROP contribute to the pathogenesis of ROP.[23,24] This supports our finding of late onset neonatal sepsis as a significant risk factor for ROP.
Other significant associations with ROP that were found in our study, although only in univariate analysis, include RDS, small for GA at birth, occurrence of hyperglycemia, NEC, PDA, and seizure. In addition, younger maternal age was also found to be associated with ROP group, but this can be related to more preterm birth in these young mothers as shown in the group of ROP neonates, and as reported by other studies. Comparison of risk factors between our study and other reports is demonstrated in Table 5.
There are few limitations in our study, mainly being a retrospective study, and second, the data are obtained from a single tertiary institution with a limited number of neonates and may not be a representative of the other NICU centers in the country. In addition, there was the exclusion of 62 neonates including half who expired during the neonatal course, and the other half being excluded because they had incomplete clinical data, or lost to follow-up, which might have effect on the true incidence rate.
ROP is still commonly encountered in neonatal practice. The current results show possible increasing incidence in Oman compared to some reports from other countries. This fact highlights the importance of prevention and treatment of ROP and its further complications. The leading risk factors for the development of ROP in our setting were prematurity, low BW, as well as prolonged oxygen therapy. Late-onset neonatal sepsis is another independent risk factors found in our cohort. Strategies to improve the quality of survival and long-term outcome includes prevention/reduction of premature births by good control and treatment of maternal illnesses, optimal oxygen use to reduce the risk of ROP, in addition to prevention of infection by strict adherence to infection control measures, early recognition and adequate treatment of sepsis. Screening and close surveillance by experienced ophthalmologists is important to diagnose and treat this common complication of prematurity and prevent subsequent visual impairment or blindness. A prospective population/regional based study is recommended to provide a more accurate data on the overall incidence and risk factors of ROP in Oman and the Middle East compared to the Western World. Meanwhile, it is important to adopt strategies to prevent ROP considering the reported international and local risk factors.
There are no conflicts of interest.