If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address correspondence to: Swati Agarwal-Sinha, MD, FASRS Pediatric Retina and Pediatric Ophthalmology, Seattle Children...s Hospital, Department of Ophthalmology University of Washington, 4800 Sand Point Way, NE, Seattle, WA 98105 United States.
Department of Ophthalmology, College of Medicine, University of Florida, Gainesville, FloridaDeparment of Pediatric Ophthalmology, Seattle Children's Hospital, University of Washington, Seattle, Washington
Premature infants, after anti–vascular endothelial growth factor injections for retinopathy of prematurity, have persistent peripheral avascular retina (PAR). PAR is ablated with laser; however, physiologic growth of the retinal vasculature in the long term has not been measured. The purposes of this study were to measure retinal vessel growth after treatment with intravitreal bevacizumab (IVB) for retinopathy of prematurity, using serial fluorescein angiography (FA), until age 3 years, and to assess the timing for providing laser ablation in PAR.
Methods
Data from an observational, longitudinal clinical study were collected. Angiographic images of eyes treated with IVB were included; imaging data from laser photocoagulation were excluded. All eyes underwent initial examination under general anesthesia with FA and photographic imaging. The retinal vessel length was measured from the temporal margin of the optic disc passing through the foveal center, and the lengths at subsequent FA were compared. To compare the changes in retinal vessel length over time in individual eyes, a paired-sample t test was performed.
Findings
FA images from 70 eyes (35 infants) treated with IVB were available. A total of 150 FA images were available for review; data from 125 images of good quality were used for analysis. The mean postmenstrual ages (PMAs) at first, second, third, and fourth FA were 66.2, 100.9, 135.1, and 180.7 weeks, respectively. The mean retinal vessel length was 14.177 mm at first FA and 13.199 mm at fourth FA (PMA range, 42...234 weeks). Retinal vascular lengths of individual eyes compared over time showed no statistically significant growth from the first FA to age 3 years. The changes in retinal vessel length from first to second FA were –0.117 ± 0.79 mm (p = 0.42; n = 30); from first to third FA, +0.060 ± 0.85 mm (p = 0.79; n = 15); and first to fourth FA, –0.404 ± 1.32 mm (p = 0.45; n = 7).
Implications
Beyond 65 weeks' PMA, no meaningful retinal vascular growth occurred after IVB up to age 3 years, guiding the timing for physicians if laser photocoagulation is being considered. Future studies are needed to address retinal growth changes in the growing eyes of infants.
Retinopathy of prematurity (ROP) is a potentially blinding disease associated with abnormal development of retinal blood vessels that occurs in premature infants.
Early Treatment for Retinopathy of Prematurity Cooperative Group. Final results of the early treatment for retinopathy of prematurity (ETROP) randomized trial.
). Although cryotherapy and laser photocoagulation were effective treatment modalities, they resulted in permanent ablation of the peripheral avascular retina, leaving the patient with a constricted visual field and an increased risk for pathologic myopia.
As the role of vascular endothelial growth factor (VEGF) in the development of ROP has been better understood, intravitreal anti-VEGF therapy has emerged as a primary treatment option for patients with ROP. In the BEAT-ROP (Bevacizumab Eliminates the Angiogenic Threat—ROP) study,
although the efficacy of anti-VEGF in inhibiting active stage 3 ROP was established, the concerns for chronic arrest of the peripheral retinal vasculature, with persistent avascular retina and reactivation of ROP as long-term complications, became a concern.
identified full vascularization, defined as vascularization that developed without an active component or clinically significant tractional elements, with treatment with anti-VEGF injection. However, more recent studies have reported much lower rates of complete retinal vascularization by identifying persistent peripheral avascular retina (PAR) more accurately using IV fluorescein angiography (FA). Chen et al
analyzed FA images from 92 eyes treated with IVB for type I ROP at 60 weeks' postmenstrual age (PMA), of which 89 eyes (96.7%) were reported as having not reached maturity, and 3 eyes (3.3%) were reported as having reached full vascular maturity. Gonzalez and Agarwal-Sinha
Fluorescein angiographic features post-intravitreal bevacizumab for retinopathy of prematurity: can they support rescue laser photocoagulation to the avascular retina.
recently reported on the angiographic findings from a series of 26 infants (47 eyes), of which all eyes showed chronic vascular arrest angiographically, until age 3 years; these findings were similar to those from the study from Lepore et al.
Follow-up to age 4 years of treatment of type 1 retinopathy of prematurity intravitreal bevacizumab injection versus laser: fluorescein angiographic findings.
Only 1 infant in the series had late peripheral tractional fibrous proliferation in the periphery with bevacizumab injection; it was diagnosed early and treated with barrier laser photocoagulation, with preservation of excellent visual outcome.
Fluorescein angiographic features post-intravitreal bevacizumab for retinopathy of prematurity: can they support rescue laser photocoagulation to the avascular retina.
The studies available to date have reported variable responses of the PAR and/or full vascularization with bevacizumab injection; however, measurements of physiologic growth of the retinal vasculature in the long term has not been reported. Two of the main questions that arise are, When can no further retinal vascular growth after IVB therapy be expected?; and What is the ideal time to perform rescue laser to the PAR? This study analyzed angiographic images from infants treated with IVB, using measurements of retinal vascular length and evaluations of growth over time, from up to four serial angiographic sessions, until age 3 years in both treated and untreated eyes.
Patients and Methods
Study Design and Patient Selection
This retrospective, longitudinal, observational study was conducted in premature infants with type I ROP treated with off-label IVB between July 2014 and February 2020. The study protocol was approved by the institutional review board at the University of Florida and was compliant with the Health Insurance Portability and Accountability Act. Data from premature infants with type I ROP (as defined using the ETROP criteria
Early Treatment for Retinopathy of Prematurity Cooperative Group. Final results of the early treatment for retinopathy of prematurity (ETROP) randomized trial.
) and treated with IVB were included in the study. Informed consent for participation was obtained from the parents of the infants. Patients with eyes with media opacity that prevented staging, and those treated with primary laser photocoagulation, were excluded. Infants treated with IVB injection (0.75 mg/0.03 mL) were followed up weekly or biweekly after treatment until 55 weeks' PMA. Follow-up was then spaced out to 4 to 5 weeks, and a complete eye examination, with depressed retinal examination, was performed in-clinic at a mean of 60 weeks' PMA. At a mean PMA of 65 to 70 weeks, first eye examination under anesthesia (EUA) was performed. Under anesthesia, depressed retinal examination using binocular indirect ophthalmoscopy was performed to assess the stage and zone of ROP and any recurrence of sight-threatening stages of ROP. Retinal color fundus images were captured using the RetCam 3 imaging system (Clarity Medical Systems, Pleasanton, California) with a 130° field of view. Digital FA was performed using a bolus of 10% fluorescein solution 0.1 mL/kg IV, followed by a saline flush. The quality of the images was reviewed by the performing senior physician (S.A.S.) and second physician (J.D.H.). The fundus images of good quality collected from each eye were used for data analysis. Data from infants who underwent laser photocoagulation at the time of image collection were excluded from the analysis. The best available images that included the temporal edge of the optic disc to the temporal vascular–avascular junction of the study eye of each patient were identified and exported in .jpeg format for analysis.
Recurrence was defined as stage 3/4/5 at any time after injection. If no recurrence of sight-threatening ROP was noted, then repeat EUA every 6 to 8 months until 3 years of age was recommended. Data collection included gestational age at birth, birth weight, postmenstrual age (PMA) at IVB treatment, recurrent stage 3 ROP status, and FA images for retinal vascular growth measurements until age 3 years. Image processing was performed in collaboration with the Department of Biomedical Engineering, University of Florida (Gainesville, Florida) (H.N., J.P., S.R.).
Retinal Vascular Growth Measurement
To examine the extent of retinal vascularization, the distance from the temporal margin of the optic disc through the fovea to the temporal vascular–avascular junction in each eye was analyzed at all follow-up eye examinations, for a total of 150 images. Figure 1 shows the anatomic landmarks of interest and an example measurement. ImageJ, (1.53g 4 December 2020) a Java-based open-source image-processing software developed by the National Institutes of Health and the Laboratory for Optical and Computational Instrumentation (NIH, Bethesda, Maryland; http://imagej.nih.gov.ij), was used to calculate this measurement. Due to the complexity of this measurement, it was deemed that computer-aided image-analysis methods would be challenging to program for accurate measurement of the distance similarly across various time points without significant validation. Prior to manual distance measurement, additional image-threshold adjustments to the brightness and contrast were needed to improve vasculature visibility. Figure 2 provides an example of the original image compared to the post-processed image. The user then manually measured the distance from the optic disc to the temporal edge. Properly converting from pixels to millimeters was essential for extracting useful measurements from these images. Oloumi et al
described using images from RetCam with the wide-angle ROP lens (130°) for computer-aided diagnosis of ROP. The images in that study were obtained at a size of 480 × 640 pixels and used spatial resolution of 0.030 mm/pixel based on the manufacturer's specifications. In the present study the same wide-angle ROP lens (130°) was used, but images were taken with higher size of 1200 × 1600 pixels for the same field of view; therefore, spatial resolution was converted accordingly by dividing 0.030 by a factor of 2.5. The spatial resolution in the present study was set to 1 pixel/0.012 mm.
Figure 1The retinal vessel length (green) is measured from the temporal edge of the optic disc (yellow) through the fovea center of the macula (blue) to the temporal vascular–avascular junction (orange).
Figure 2Sample fluorescein angiography of the retina for analyzing vessel length. The original image (A) was edited with increased brightness and contrast to improve vasculature visibility (B).
The mean (SD) and range of retina vessel length measurements were calculated for each of four FA sets and for the overall data. In order to compare change in retinal vessel length over time in individual eyes, a paired-sample t test was performed using those with quality images available for measurement. For all tests, the α level was 0.05, and the CI was set at 95%. Statistical analysis was performed using Excel (Version 2020) software (Microsoft Corporation, Redmond, Washington) and SPSS (SPSS 28 - 2021) software (SPSS Inc, Chicago, Illinois).
Results
A total of 70 eyes (35 infants) with type I ROP underwent FA imaging. A total of 63 eyes were treated with IVB; 7 eyes did not require treatment. Of the 63 treated eyes, 36 eyes (57.1%) had ROP in zone 1; 27 (42.9%), in zone 2, at the time of treatment. The mean PMAs at first, second, third, and fourth FA were 66.2, 100.9, 135.1, and 180.7 weeks, respectively; n = 46 [65.7%], 22 [31.4%], and 12 [17.1%] eyes at second, third, and fourth FA).
The mean gestational ages and birth weight were 24.7 weeks PMA and 702.9 g, respectively. ROP was diagnosed at a mean PMA of 32.7 weeks, and first IVB treatment was given at a mean of 36.1 weeks' PMA. Recurrence requiring additional IVB occurred in 7 of 63 eyes (11.1%) at a mean of 44.0 weeks' PMA and none after 49 weeks' PMA. Initial outpatient screening ended at a mean of 57.6 weeks' PMA, with last follow-up (either in clinic or EUA) occurring at a mean of 134.3 weeks (range, 64–246 weeks) (Table I).
Table IClinical characteristics of infants (N = 35) and eyes (n = 63) with type I ROP treated with IVB.
All 70 eyes underwent initial EUA with FA and RetCam imaging at a mean of 66.2 weeks' PMA. First FA images were obtained at a mean of 30.1 weeks after IVB treatment. Second FA was obtained in 46 eyes (65.7%), third FA in 22 eyes (31.4%), and fourth FA in 12 eyes (17.1%). In total, 150 FA sessions were included for review, and 125 of 150 images (83.3%) were determined to be of sufficient quality for further analysis (Table II). The mean retinal vessel length at first FA was 14.177 ± 1.044 mm (66.2 ± 11.3 weeks' PMA); second FA, 14.072 ± 0.971 mm (100.9 ± 25.3 weeks' PMA); third FA, 13.597 ± 0.980 mm (135.1 ± 36.4 weeks' PMA); and fourth FA, 13.199 ± 0.785 mm (180.7 ± 16.4 weeks' PMA).
Table IIRetinal vessel length measured using FA at four time points.
No. of FA
Good Eyes, n/N (%)
Age, mean ± SD (range), wk PMA
Retinal Vessel Length, mean ± SD (range), mm
First
55/70 (78.6)
66.2 ± 11.3 (42–94)
14.177 ± 1.044 (10.409–16.113)
Second
40/46 (87.0)
100.9 ± 25.3 (69–190)
14.072 ± 0.971 (12.017–15.760)
Third
19/22 (86.3)
135.1 ± 36.4 (98–234)
13.597 ± 0.980 (12.152–15.247)
Fourth
11/12 (91.7)
180.7 ± 16.4 (161–206)
13.199 ± 0.785 (12.209–15.017)
Overall
125/150 (83.3)
96.0 ± 41.2 (42–234)
13.969 ± 1.028 (10.409–16.113)
FA = fluorescein angiography; PMA = postmenstrual age.
In order to compare individual eyes for changes in retinal vessel length over time, three separate paired t-test analyses were performed to compare each subsequent FA to the baseline FA image (Table III). A longitudinal trend analysis across all four time points could not be performed due to limited number of eyes that had examinations with images of sufficient quality across all four time points. At each subsequent time point, fewer patients had reached the age for another examination, and not every examination provided a quality image; therefore, a limited the number of eyes could be included at each time point. Separate paired t-test analyses allowed for the comparison of individual eyes at each time point back to baseline measurements and maximized analysis of available eyes with sufficient image quality at each time point. The change in vessel length on the second FA compared with first FA (n = 30) was –0.12 ± 0.79 mm (95% CI, –0.42 to +0.18; p = 0.42); third FA versus first FA (n = 15), +0.06 ± 0.86 mm (95% CI, –0.41 to +0.53; p = 0.79); fourth FA versus first FA (n = 7), –0.40 ± 1.32 mm (95% CI, –1.63 to +0.82; p = 0.45). The results are displayed in Figure 3. No statistically significant change in retinal vascular growth was noted from the first FA performed at a mean of 66.2 weeks' PMA to any of the subsequent time points when individual eyes were compared until 3 years of chronologic age.
Table IIIInterval changes in retinal vessel length compared to first fluorescein angiography (FA).
In a subanalysis, the mean retinal vascular length at first FA in infants treated for zone 1 ROP (14.180 ± 1.187 mm) was compared to that in infants treated for zone 2 ROP (14.173 ± 0.810 mm) and found to be similar, with no statistical difference (mean difference, +0.007 mm; 95% CI, –0.574 to +0.589; P = 0.98) (Table IV).
Table IVRetinal vessel length at first FA in patients with zone 1 versus zone 2 ROP at time of IVB treatment.
Six of 7 eyes with recurrent stage 3 ROP requiring an additional IVB treatment had quality imaging available from the first FA and were used for analysis. Eyes with recurrent ROP had a shorter mean retinal vessel length (13.470 ± 1.741 mm) at first FA compared to those that received only one IVB treatment (14.241 ± 0.922 mm), but this was not statistically significant (P = 0.09) (Table V).
Table VRetinal vessel length at first FA in patients with versus without ROP recurrence requiring additional IVB treatment.
Seven of 63 eyes had recurrent stage 3 ROP and received repeat bevacizumab injection. Six of 7 re-treated eyes had quality imaging available from first FA.
)
2
13.470 ± 1.742
FA = fluorescein angiography; IVB = intravitreous bevacizumab; ROP = retinopathy of prematurity.
Two-tailed t test.
† Seven of 63 eyes had recurrent stage 3 ROP and received repeat bevacizumab injection. Six of 7 re-treated eyes had quality imaging available from first FA.
Of the 7 eyes that did not receive treatment, there were 15 quality images of 21 total FA sessions (71.4%) across the four time points, with a mean retinal vessel length of 14.33 ± 0.85 mm, which was similar to that seen in the treated eyes. Given that the number of images in the nontreated eyes was low, we could not compare individual eyes for changes in retinal vessel length over time.
Discussion
In this series of premature infants treated with IVB for type I ROP, 70 eyes of 35 infants were included to study the growth of retinal vasculature starting from the mean age of 66 weeks' PMA until age 3 years. These study data showed the lack of retinal vessel growth after IVB in both the treated and untreated eyes when patients were followed up until age 3 years. Prior studies have reported on the continued development of peripheral retinal vessels to full vascularization in eyes treated with IVB,
unlike conventional laser photocoagulation which can lead to the destruction of the peripheral retina.
One hypothesis made in the studies was that the arrest of the retinal vascular growth was secondary to the use of the anti-VEGF treatment. Similar comparable vascular changes on FA in untreated premature infants were noted by Purcaro et al,
hypothesized that bevacizumab does not induce regression of ROP, but rather seems to "freeze" ROP in the acute involutional phase. It was hypothesized that the retina-related findings at the vascular–avascular junction were the result of oxygen-induced arrest of vascular development that occurred during phase 1 of development (22–30 weeks' PMA) rather than being the result of anti-VEGF treatment.
Fluorescein angiographic features post-intravitreal bevacizumab for retinopathy of prematurity: can they support rescue laser photocoagulation to the avascular retina.
In the past, without the use of injections, the avascular retina was ablated by laser and or cryotherapy, so no avascular retina was noted. Similar abnormal vascular patterns on FA were seen at a similar prevalence after spontaneous regression of ROP.
These findings suggest that the abnormal vascular patterns identified by FA in patients with ROP result from the disease process itself rather than resulting from exposure to anti-VEGF treatment.
With FA taking preeminence in defining persistent retinal vasculature changes after IVB until age 4 years,
Follow-up to age 4 years of treatment of type 1 retinopathy of prematurity intravitreal bevacizumab injection versus laser: fluorescein angiographic findings.
A recent longitudinal angiographic imaging study, in which 15 of 26 infants (47 eyes) with type I ROP treated with IVB monotherapy until age 3 years, described the angiographic features to assist in the low risk for recurrent sight-threatening retinopathy (2%), with no rescue laser photocoagulation to the PAR in 98% of eyes. Only 1 eye in the series
Fluorescein angiographic features post-intravitreal bevacizumab for retinopathy of prematurity: can they support rescue laser photocoagulation to the avascular retina.
had late-onset peripheral fibrous proliferation at 16 months of chronologic age, which was diagnosed early because of the approach of frequent EUA beyond age 1 year and timely treatment with barrier laser photocoagulation, and with excellent preservation of vision, unlike the late-recurrence patients reported in other studies with poor visual outcomes.
described 39 eyes (43.8%) with avascular arrest, 34 eyes (38.2%) with vascular arrest and persistent tortuosity, 16 eyes (18%) with ROP reactivation, and 3 eyes (3.3%) with complete vascular maturity to within 2DD from the ora serrata based on the findings by Fierson
AAP American Academy of Pediatrics Section on Ophthalmology, AAP American Academy of Ophthalmology, AAP American Association for Pediatric Ophthalmology and Strabismus, AAP American Association of Certified Orthoptists. Screening examination of premature infants for retinopathy of prematurity.
and the joint statement screening guideline criteria. The authors performed an EUA at 60 weeks' PMA with FA to measure the area of ischemia or PAR in the three groups prior to laser photocoagulation to the avascular retina.
While the other studies have quantitatively or qualitatively assessed the degree of PAR, no available studies have directly measured the length of retinal vasculature using FA images or its growth over time. In this study, FA images from patients with ROP treated with IVB were analyzed, and the retina vasculature was measured from the temporal margin of the optic disc passing thru the foveal center to the temporal vascular–avascular junction, and compared over time to up to four total FA sessions, starting at mean of 66.2 weeks' PMA at different time points until age 3 years to determine whether the vessels mature to the ora serrata. The data showed that all treated eyes (100%) maintained PAR at all time points until age 3 years. The mean retinal vessel length at first FA was 14.18 mm; second FA, 14.07 mm; third FA, 13.60 mm; and fourth FA, 13.20 mm. Individual eyes were compared over time using three separate paired t-test analyses, one for each subsequent FA (second, third, and fourth) compared to the first FA image. Each of these time points showed no statistically significant change in retinal vessel length over time compared to baseline. Although the PAR was not being directly measured, the data suggest that the avascular retina did not decrease in size over time.
In the present study, recurrence of staged disease requiring additional IVB treatment occurred in 7 of 63 eyes (11.1%) at a mean of 8 weeks after initial treatment, and none occurred after 49 weeks' PMA. The recurrence rate was similar to the findings of Mintz-Hittner et al,
who reported a prevalence of recurrence of 8.3% (20/241) in infants and 7.2% (34/471) in eyes treated with IVB. Patients with recurrence in the present study had a lower mean vessel length (13.47 mm) at initial FA compared to those without recurrence (14.24 mm), but this finding was not statistically significant (P = 0.09). Even the nontreated eyes (n = 7) in the series showed a mean retinal vessel length of 14.33 ± 0.85 mm, which was similar to that in the treated eyes with PAR.
The findings from the present study imply that beyond 65 weeks' PMA, no meaningful additional retinal vascular growth occurs after treatment with anti-VEGF for sight-threatening ROP in either treated or untreated eyes when measured until age 3 years. Although the data suggest no further retinal growth, the mean retina length seems to get smaller with each time point. First, this finding may have been due to the fact that the later time points had smaller sample sizes, and the earlier time points were skewed higher by eyes with longer vessel lengths. Second, and most important, the optical effects of the growing infant eye may affect the measurement of the retinal vessel growth. The axial length (AL) measurement in a full-term (40 weeks) neonatal eye is reported as 16.6 to 17 mm. Summing up studies on the AL in premature eyes, it was concluded that premature eyes with advanced threshold retinopathy had smaller eyes, with a weekly axial elongation prior to term, ranging from 0.13 to 0.23 mm/wk.
measured the AL in a pediatric population with no intraocular pathology in 330 eyes of 165 patients. The study reported AL measured in different age groups starting from age 3 months until age 7 years. The mean AL at 3 to 6 months was 19.76 mm, with the steepest increase in AL noted in the first 10 months of life to 20.56 mm. Mutti et al
noted similar findings, with mean ALs of 19.03 and 20.23 mm at 3 and 9 months, respectively.
The AL measurements from 10 eyes in the present series are shown in Table VI. If the chronologic ages of the patients were adjusted to the corrected ages of a full-term infant, the mean AL at each time point for FAs 1 to 4 based on the current literature can be averaged to 19.7643 mm at FA 1 for age <1 year, 21.072 mm for ages 1 to 3 years, 21.75 mm for ages 2 to 4 years, and 21.75 mm for ages 3 to 4 years (Table VI). Even when the AL is adjusted for the corrected age, the retinal growth measurements likely cannot be mathematically adjusted for the following reasons.
Table VIMean axial length for corrected age at fluorescein angiography.
Eye growth involves not only the increase in AL but also the change in eyeball size and shape, which includes AL (defined as the length from the posterior corneal surface to the retinal surface), horizontal width (transverse), and vertical height of the internal eye (sagittal) along the cardinal axis. The average size of the human adult eye is 24.2 mm horizontal (transverse) × 23.7 mm vertical (sagittal) × 22 to 24.8 mm axial. The transverse diameter of the eyeball varies from 21 to 27 mm. In a study by Lim et al,
the eye size and shape were measured in a full-term neonate by magnetic resonance imaging (MRI). Oblateness of the eye was calculated as 1 − (AL/width) or 1 − (AL/height). The study reported that eyes with longer ALs had statistically significantly greater widths, heights, volumes, and surface areas than did eyes with shorter ALs. The investigators concluded that in their series of 346 eyes from 173 Singaporean Asian neonates, most were born with prolate eyes. In another similar study in 105 patients aged 1 month to 19 years, changes in eyeball shape were evaluated using MRI. In patients aged 1 month to 6 years, the oblate to prolate changes in shape were significantly correlated with age. In patients aged 7 to 19 years, it was significantly correlated with spherical equivalent refraction.
developed software to calculate the shape of the retinal surface in emmetropic and myopic eyes at age 18 to 36 years by MRI using the anteroposterior, transverse, and vertical (sagittal) measurements. The investigators noted that all emmetropic retinas were oblate in shape in transverse, axial, and sagittal sections,
meaning that the retinal surfaces steepened away from the posterior retinal vertex in all meridians and had more width and height than length. Considering transverse and/or sagittal lengths of the eye—which are required, along with the axial length, to address the change in retinal and eye shape and size over time—were unavailable in these premature infants, it was not possible to mathematically adjust the retinal growth measurements in the present study.
It is known that the dose of anti-VEGFs balances the effect of regression of neovascularization against the development of normal retinal vascularization, with higher doses inhibiting the normal retinal angiogenesis. Findings from studies in animal models suggest that a precise titration of anti-VEGF dose to balance inhibition of neovascularization with retinal nonperfusion and longer observation after anti-VEGF treatment is required for evaluation of its effect on vascular recovery and physiologic angiogenesis.
reported on the efficacy of aflibercept in reducing areas of nonperfused retina and its effect on normal angiogenesis in oxygen-induced retinopathy in a mouse model of ROP. In the short term (3 days after injection), the neovascularization was regressed, but no difference in nonperfusion was noted. A significant difference in the reduction of retinal nonperfusion was noted in the long term (11 days after injection) in the mouse model. In addition, optimal doses led to vascular recovery, whereas higher doses inhibited normal angiogenesis. The findings from the mouse model were similar to those from the study done in Beagle pups by Lutty et al.
Pediatric Eye Disease Investigator Group Assessment of lower doses of intravitreous bevacizumab for retinopathy of prematurity: a phase 1 dosing study.
a 0.75-mg dose of bevacizumab was used for treatment, which can be considered high when in a recent Phase I dosing study by the PEDIG (Pediatric Eye Disease Investigators Group) group, a bevacizumab dose as low as 0.031 mg and a very low dose of 0.004 mg, were noted to be effective in the treatment of ROP. But the PEDIG studies had short-term efficacy end points, with success defined at 4 weeks after treatment, and with 25 (41%; 61 eyes) receiving additional treatment. The recurrence rate was higher in the PEDIG study, with the majority treated with laser photocoagulation. Given that the patients enrolled in the PEDIG study were not followed up to age 3 years with bevacizumab monotherapy, the long-term effects of bevacizumab on retinal vascular angiogenesis in infants were unavailable.
Pediatric Eye Disease Investigator Group Assessment of lower doses of intravitreous bevacizumab for retinopathy of prematurity: a phase 1 dosing study.
Doses of 0.050, 0.625, and 0.75 mg of bevacizumab have been associated with findings of persistent avascular retina. The risk for recurrence was lower without rescue laser photocoagulation to the avascular retina with follow-up until 3 years of age.
Fluorescein angiographic features post-intravitreal bevacizumab for retinopathy of prematurity: can they support rescue laser photocoagulation to the avascular retina.
Follow-up to age 4 years of treatment of type 1 retinopathy of prematurity intravitreal bevacizumab injection versus laser: fluorescein angiographic findings.
Comparison of fluorescein angiographic findings in type 1 and type 2 retinopathy of prematurity with intravitreal bevacizumab monotherapy and spontaneous regression.
In the present series, EAU was repeated for depressed retina and late recurrence of stage 4/5 ROP that had been reported in the literature, with poor visual outcomes.
In the present series, 1 eye in an infant with late recurrence of early peripheral stage 4 ROP was diagnosed on EUA at 15 months of age and treated with barrier laser, with excellent visual outcome.
The long-term impact of general anesthesia on the developing brain in infants and children, including neurotoxicity, is a pressing concern. Findings from preclinical studies in animal models have suggested that early-life exposure to general anesthesia can lead to neurotoxicity with lasting behavioral and cognitive deficits
In December 2016, based on the available preclinical and clinical data, the US Food and Drug Administration issued a warning that "repeated or lengthy use of general anesthesia and sedation drugs during surgeries or procedures in children younger than 3 years or in pregnant women during their third trimester may affect the development of children's brains #50. Page 906. (www.fda.gov/Drugs/DrugSafety/ucm532356.htm)." Lengthy is defined as >3 hours. This warning, issued by the Food and Drug Administration without new or compelling evidence in humans, also states, “Additional high quality research is needed to investigate the effects of repeated and prolonged anesthesia exposures in children, including vulnerable populations #50. page 907.”
In the recent multicentric clinical trials GAS (General Anesthesia vs Spinal Anesthesia) and PANDA (Pediatric Anesthesia and Neurodevelopment Assessment), neurodevelopmental testing revealed a string of evidence that just under an hour of general anesthesia in early infancy was not associated with poorer neurodevelopmental outcomes.
Neurodevelopmental outcome at 2 years of age after general anaesthesia and awake-regional anaesthesia in infancy (GAS): an international multicentre, randomised controlled trial.
GAS Consortium. Neurodevelopmental outcome at 5 years of age after general anaesthesia or awake-regional anaesthesia in infancy (GAS): an international, multicentre, randomised controlled equivalence trial.
The MASK (Mayo Anesthesia Tolerability in Kids) study supports that multiple, not single, exposures to procedures requiring general anesthesia before the of age 3 years were associated with a specific pattern of neurologic deficits that may have effects on learning and behavior, but the study did not define the factors that predict who will develop the deficits. Andropoulos and Greene
stated, “Until reassuring new information from well-designed clinical trials is available, we are concerned that the FDA warning will cause delays for necessary surgical and diagnostic procedures that require anesthesia, resulting in adverse outcomes for patients 50. page 906.” This area of ongoing research is focused on preclinical science and meaningful translation into the clinical care of infants and children to provide safe anesthesia and to study its consequences leading to neurotoxicity.
In the present series, EUA was brief (≈15–30 minutes; <1 hour) and, when possible, was combined with other surgical procedures to minimize the need for multiple administrations of anesthesia. Premature infants who are treated for sight-threatening ROP may have multiple associated comorbidities that require frequent long anesthesia for life-saving surgeries for necrotizing enterocolitis, hydrocephalus, intraventricular bleeds, and congenital heart defects. The benefits of brief anesthesia for a depressed eye EUA in infants treated with anti-VEGF injections outweighs the risk for late tractional detachment and blindness.
There were a few limitations in the study. First, the transverse and/or sagittal lengths of the eye, as well as the axial length, were unavailable in these premature infants, so the change in retinal and eye shape and size over time could not be addressed, and the retinal growth measurements could not be mathematically adjusted with the current data to address the optical effects of the growing eye. Second, EUA was done with concerns for late recurrence after treatment with IVB, which is not common practice among providers. Third, 16.7% of the images could not be analyzed either because the images were of poor quality or did not include all of the necessary structures for measurement. The relatively short window after injection of the fluorescein dye makes this aspect of the image collection difficult, especially when quality images of both eyes are required. While the overall number of eyes at first FA was relatively high, there was a decreased number of patients available from each successive FA. There were 70 eyes at first FA but only 12 eyes at fourth FA. This decrease was due to many of the patients involved in the study having not yet reached the recommended age for additional FA. This reduced number of patients limited the power of the analysis of data from the latter time points, as only 7 eyes were included for analysis from the fourth FA, while 30 eyes were able to be included from the second FA.
Anti-VEGF treatment plays an important role in the treatment of type I ROP. While it avoids the complications of permanent ablation of peripheral avascular retina caused by conventional treatments, most patients do not achieve full vascular maturity. The study highlights that the retinal vascular length after bevacizumab monotherapy remained similar in individual eyes throughout four time points beginning at 66 weeks' PMA until 3 years of chronologic age, suggesting no further growth of retinal vasculature beyond 66 weeks' PMA, which was also seen in the untreated eyes. The persistence of chronic vascular arrest and the lack of normal anterior retinal growth after treatment with anti-VEGF therapy justify a longitudinal study of dose titration and longer follow-up in infants with ROP to study vascular growth and recovery through adulthood, along with the other optical parameters that may affect the size and shape of the growing eye.
Conclusions
The findings from the present study imply that beyond 65 weeks' PMA, no meaningful additional retinal vascular growth occurs after anti-VEGF therapy for sight-threatening ROP when measured until age 3 years, even in untreated eyes. These findings are clinically valuable as the degree of PAR temporally may remain the same, guiding the timing for physicians if laser photocoagulation is being considered for PAR.
DECLARATION OF INTEREST
The authors have indicated that they have no conflicts of interest with regard to the content of this article.
Acknowledgments
JD Hammer and Henry Nguyen both authors contributed equally to this manuscript. J. D Hammer, MD, contributed to literature search, figures, data analysis, and drafting the manuscript. Henry Nguyen, contributed to literature search, figures, data analysis, and drafting the manuscript. Jacqueline Palmer, contributed to figures, data analysis and drafting the manuscript. Sarah Furtney, PhD, contributed to data analysis and editing of the manuscript. Swati Agarwal-Sinha, MD, FASRS, contributed to study design, data collection, editing and writing and final approval of the manuscript.
Role of Funding Source
This study received no funding.
References
Good WV
Hardy RJ
Dobson V
et al.
The incidence and course of retinopathy of prematurity: findings from the early treatment for retinopathy of prematurity study.
Early Treatment for Retinopathy of Prematurity Cooperative Group. Final results of the early treatment for retinopathy of prematurity (ETROP) randomized trial.
Fluorescein angiographic features post-intravitreal bevacizumab for retinopathy of prematurity: can they support rescue laser photocoagulation to the avascular retina.
Follow-up to age 4 years of treatment of type 1 retinopathy of prematurity intravitreal bevacizumab injection versus laser: fluorescein angiographic findings.
AAP American Academy of Pediatrics Section on Ophthalmology, AAP American Academy of Ophthalmology, AAP American Association for Pediatric Ophthalmology and Strabismus, AAP American Association of Certified Orthoptists. Screening examination of premature infants for retinopathy of prematurity.
Comparison of fluorescein angiographic findings in type 1 and type 2 retinopathy of prematurity with intravitreal bevacizumab monotherapy and spontaneous regression.
Neurodevelopmental outcome at 2 years of age after general anaesthesia and awake-regional anaesthesia in infancy (GAS): an international multicentre, randomised controlled trial.
GAS Consortium. Neurodevelopmental outcome at 5 years of age after general anaesthesia or awake-regional anaesthesia in infancy (GAS): an international, multicentre, randomised controlled equivalence trial.
The information in this article was previously presented at: the Association for Research in Vision and Ophthalmology Annual Meeting [virtual], May 1–7, 2021; the American Society of Retina Specialist Annual Meeting, San Antonio, Texas, October 8–12, 2021; and the European Society of Retina Specialists Annual Meeting [virtual], September 10–11, 2021.
Two-tailed paired sample t test.
00407-6&id=gr1.jpg){kind=link}
00407-6&id=gr2.jpg){kind=link}
00407-6&id=gr3.jpg){kind=link}