This cross sectional study was done in 2021–2022 on 4–6 years old preterm children who visited the ROP clinic of Nikookari Hospital in Tabriz, Iran between 2015–2017 for the diagnosis and treatment of ROP. In this study 79% of eyes receiving bevacizumab had Stage III ROP and 26% Stage I. While in children with spontaneously improved ROP, only 6% had ROP in Stage III and 61% in Stage I. This difference is significant and can justify the need of patients to receive bevacizumab. It was found that 21% of people in the group receiving bevacizumab had their ROP involvement in Zone I and 70% in Zone II, but in the group with spontaneous improvement of ROP, there were no cases of Zone I involvement and 45% of Zone I involvement. II and 55% had III Zone involvement.
MACULA:
According to this study, the average thickness of the central 0.5 mm of the macula (fovea) was 235.46 ± 11 µm in the group receiving bevacizumab, 267.70 ± 12 µm in the group with spontaneous improvement of ROP, and in the group with no ROP, it was 269.48 ± 10 micrometers. According to these findings, the mean thickness of the fovea in the group receiving bevacizumab was significantly lower than the other two groups which can show the negative effects of bevacizumab on the normal macular development in these infants.
Among factors affecting foveal thickness, macular edema has been reported in approximately 50% of sever ROP cases [11, 12] and it has been reported that the prevalence of edema increased with an increase in ROP stage.[11] In infants having macular edema, the fovea and inner nuclear layer thickness increased with edema severity, variations in layer thicknesses in infants at 36 ± 1 weeks' postmenstrual age may reflect abnormalities resulting from delay in foveal development that may be impacted by macular edema, ROP, or both [12]. In our study time point of visit was different at age of 4 to 6 year and in these children no macular edema was detected it can be due to the period of time passed.
In a study by lepore and et al Among the bevacizumab-treated eyes available at 4 years of age, all showed abnormalities at the periphery (avascular area, vessel leakage, shunts, abnormal vessel branching, and tangles) or the posterior pole (hyperfluorescent lesions, absence of foveal avascular zone). These lesions were not observed in the majority of the lasered eyes.[13]. But they did not investigate foveal thickness in their study.
vinekar et al investigated macular changes in acute ROP in infants. Obtaining OCT from infants under sedation. (29.1%) with stage 2 ROP showed abnormal foveal changes despite clinically normal foveae. The mean central foveal thickness was 206.5 _+ 98.7 for stage2. They described 2patterns of foveal involvement “pattern A,” which was characterized by dome shaped foveal elevation and cystoid spaces with highly reflective intervening vertical septae, and “pattern B,” which was characterized by preservation of the foveal depression with fewer intraretinal cystoid spaces. At 52 weeks’ postmenstrual age all of them revealed normalization of foveal contours. They suggest that transient foveal changes at the critical time of fovealization in premature infants may influence their visual acuity in the adult life [14]. Since they investigated macular changes in acute ROP increased foveal thicknesses could be possible.
In a study by villagas and et al children treated with laser had lower foveal thicknesses and lower visual acuity in comparison with spontaneously improved group and it might be due to higher stages of ROP in the laser treated eyes. They also reported that 64% had good visual acuity despite an abnormal foveal depression in 91% of total eyes, implying that an absent foveal pit and retention of the inner retinal layers does not preclude the development of good visual acuity. They confirms that foveal photoreceptor maturation can occur despite an abnormal foveal depression, which might explain the preserved visual function in many of these patients [15]
OCT on older individuals with historically mild ROP was reported by Hammer et al. They observed neurovascular abnormalities in 77.8% of cases. Years after suffering from mild ROP, their cases presented with degraded best corrected acuity attributed to mild optical aberration or metabolic effects on neural cells that are sensitive to contrast. It was also pointed out that the loss of foveal cones or increased cone–cone spacing were responsible for these changes. It was noted that there was not lack of cones years later.[16] We think that the transient edema may contribute to abnormalities in cone packing without affecting the actual number of cones by causing increased physical separation between adjacent cones that may inhibit their tight packing during time of immaturity [14, 16].
Former studies have demonstrated that children having regressed ROP had dysplasia of the central foveal structure, including a shallow or absent foveal pit, a thicker outer nuclear layer, and a inner retinal layer overlaying the fovea shown by OCT. and also foveal microvasculature may play a role in foveal pit formation. Maybe the mechanical stress or the ingrowing vessels cause the failure in foveal pit formation.[17, 18]
In a study by vogel and et al was shown that Intravitreal bevacizumab therapy for ROP is associated with more rapid outer retinal thickening at the foveal center, whereas LPC (laser photocoagulation) is associated with earlier extrusion of the inner retinal layers and delayed development of the EZ at the foveal center.[19]
In a study by kizilay and et al Central foveal thickness (CFT) was significantly higher in infants who had treatment for ROP. Parafoveal and perifoveal retinal thicknesses in certain quadrants were higher in those receiving LPC treatment than in others. CFT was significantly lower in group spontaneously improved than in both groups, bevacizumab and laser (p < .001). The parafoveal average, temporal, inferior, and nasal thickness values and the perifoveal average, superior, and temporal thickness results were significantly higher in group laser than in both other groups [20] actually its inevitable to have higher thickness in laser group but in contrast to our study CFT was lower in spontaneously improved group rather than in the bevacizumab group.
RNFL:
This study indicates a significant difference in the thickness of the RNFL between the group receiving bevacizumab and the other two groups, so that in the group receiving bevacizumab, the thickness of the nerve layer of the retina is less than the other two groups. In a study by park and et al Global average, nasal, and superior disc RNFL thickness profiles were significantly smaller in preterm children compared with full-term children [21] it is unclear if ROP affects RNFL thickness directly, or the development of ROP and laser treatment for ROP could induce more RNFL thinning .[22] As RNFL thickness around the optic nerve has been said to be thinned with diabetic retinopathy we can suggest that severe ROP could damage the axons of ganglion cells in a similar way [23] this can explain the reason of RNFL thinning in our study too. Also in a study by kopic Retinal Nerve Fiber Layer Thickness in glaucoma patients treated with multiple intravitreal Bevacizumab injections was investigated. As we know Patients with primary open angle glaucoma (POAG) also have retinal nerve fiber layer (RNFL) thinning. They wanted to know whether administering bevacizumab to patients with POAG leads to additional reduction of RNFL thickness which their Study results showed a decrease of RNFL in both groups of patients [24] so we can claim that RNFL thinning could be due to ROP itself and bevacizumab injection both.
Also it was found that in our study the thickness of the RNFL in the temporal quadrant was lower than the other quadrants and close to the nasal quadrant and it was higher in the superior quadrant than the other quadrants. In a study by hammer et al the thickness of the inner retinal layers in subjects with ROP is slightly greater nasally than temporally[16] which was slightly parallel to our study and in a study by achim and et al Patients with severe ROP had a thicker RNFL in the temporal sector compared with normal full-term controls [25]. Another explanation for the thinning of the RNFL in the quadrants except for the temporal is possible artifacts from the vessels. It has been showed that children with low birth weight have narrower arterioles and arteriolar caliber is positively correlated with RNFL thickness and macular thickness. Therefore, narrower vessels in preterm children compared with normal full-term controls could induce thinned RNFL measured by OCT [25, 26]. Some studies state the association between peripapillary RNFL distribution and the presence of severe intraventricular hemorrhage[25] Wang et al reported that peripapillary RNFL thickness on the temporal side of the disc is related with visual acuity. Peripapillary RNFL is closely correlated with multiple visual deficits including visual field defect and low-contrast sensitivity.[26]
The main strength of this study is that we investigated macular and RNFL thickness in bevacizumab treated eyes with history of ROP which was not directly measured in previous studies and also compared with spontaneously improved eyes. We also can claim that bevacizumab can negatively affect the normal macular development. The findings of this study also allow us to extend our understanding of the development of the premature fovea to hypothesize the possible clinical and long-term effects of these abnormal changes on adulthood visual acuity.
One of limitations of this study was that we could not investigate axial length and adjust it with visual acuity of participants and also children not cooperating in visual acuity exams. And also the laser treated eyes was better to enter the study as another group which was not possible for this study due to inadequate laser treated eyes.
We recommend comparison of laser treated eyes in future designed studies with these groups in order to extend our understanding of the development of the premature fovea and other aspects of ROP decreasing preventable blindness from this condition.