- Email: [email protected]
Atlas of Fluorescein Angiographic Findings in Eyes Undergoing Laser for Retinopathy of Prematurity
Atlas of Fluorescein Angiographic Findings in Eyes Undergoing Laser for Retinopathy of Prematurity Domenico Lepore, MD,1 Fernando Molle, MD,1 Monica M. Pagliara, MD,1 Antonio Baldascino, MD,1 Carmine Angora, MD,1 Maria Sammartino, MD,1 Graham E. Quinn, MD, MSCE2 Purpose: We sought to examine the clinical features of severe retinopathy of prematurity (ROP) using fluorescein angiography (FA). Design: Retrospective case series of eyes with severe acute-phase ROP that underwent FA at the time of laser photocoagulation. Participants: We included 22 eyes of 11 infants that developed ROP stage 3 in zone 1 with plus disease, 8 eyes of 4 infants classified as ROP stage 3 in zone 1 without plus disease, and 21 eyes of 11 infants that developed ROP stage 3 in zone 2 with plus disease. All eyes underwent laser photocoagulation. A total of 51 sets of digital images including FA were obtained immediately before treatment. Methods: RetCam (Clarity, Pleasanton, CA) fundus images and video digital FAs were performed under general anesthesia right before laser treatment. A 10% solution of fluorescein was intravenously administered as a bolus at a dose of 0.1 ml/kg, followed by an isotonic saline flush. Main Outcomes Measures: Fluorescein angiograms were examined retrospectively to catalog different retinal and choroidal findings Results: In eyes with severe ROP, FA clearly shows extreme variability in both retinal circulation and choroidal filling pattern. Different patterns of vessels branching at the junction between vascular and avascular retina (V–Av junction) are noted. Posterior to the V–Av junction, hypoperfused retinal areas with or without hyperfluorescent “cotton-wool-like” or “popcorn-like” lesions due to dye leakage are documented by FA. Focal dilatation of capillaries, capillary tufts formations, and rosary-bead-like hyperfluorescent lesions inside the vessels were seen; sometimes all 3 are noted. Various macular abnormalities are noted including absence of foveal avascular area and significant exudative component. Conclusions: Fluorescein angiography was useful to distinguish the deceptively featureless zone 1 junction between the vascularized and nonvascularized retina. Further studies are needed to understand the role of vascular abnormalities observed in zone 1 vascularized retina. Financial Disclosure(s): The authors have no proprietary or commercial interest in any of the materials discussed in this article. Ophthalmology 2011;118:168 –175 © 2011 by the American Academy of Ophthalmology.
Retinopathy of prematurity (ROP) is a disease of developing blood vessels in the retina of the premature infant. In 1984, an International Committee of ROP experts developed a classification (International Classification for ROP [ICROP]) characterized by the location, stage, and extent of the disease based on the ophthalmoscopic findings.1,2 In its revisited form,3 ICROP has been the common language used not only to manage ROP clinically,4 –7 but also to provide a framework for studying basic mechanisms.8,9 Based on ophthalmoscopic appearance of ROP in zones 1 and 2, Flynn and Chan-Lin10 hypothesized 2 distinct mechanisms in the pathogenesis of ROP: In particular, the authors suggested a correlation between zone I ROP features and vasculogenic process in the posterior retina. To better illustrate their hypothesis, the authors cited previous fluorescein angiographic (FA) studies of ROP.11–13 These pioneering studies clearly showed that FA, although technically challenging, could represent a major tool to study the mechanism of vessel formation and maturation in the pre-
168
© 2011 by the American Academy of Ophthalmology Published by Elsevier Inc.
mature retina. Schulenburg and Tsanaktsids14 in 2004 used FA to investigate vascular morphology in 6 cases of extremely low– birth-weight ROP. This study showed a different ROP morphology without classic shunt formation and a poorly developed capillary bed in already vascularized retina. In the late 1990s with the advent of digital imaging of the eyes of infants at risk for ROP, FA became easier and safer to perform in the neonatal intensive care unit using such instruments as the RetCam (Clarity, Pleasanton, CA). In 2006, Ng et al15 demonstrated that clear angiograms can be obtained as part of ROP screening. In the same year, we also presented preliminary results of FA in zone I ROP (Pagliara MM, Lepore D, Baldascino A, et al. Fluorescein angiographic findings in zone I ROP [abstract]. Invest Ophthalmol Vis Sci 2006;47:E-Abstract 5302). Since then, other groups have reported small series of FA findings in ROP.16,17 Building on these studies, we have conducted FA in eyes with severe ROP that are undergoing laser photoablation to examine systematically the vascular abnormalities of such eyes. ISSN 0161-6420/11/$–see front matter doi:10.1016/j.ophtha.2010.04.021
Lepore et al 䡠 FA in Severe ROP Table 1. Circulation Patterns: Number of Eyes (%) in our Series Showing Arterial Phase Appearance Delayed up to 30 Seconds, Venous Phase Appearance Delayed up to 1 Minute, and Abnormal Choroidal Filling (as described in Fig 1A, B)
Branching
Choroid
Arterial Phase
Venous Phase
Delayed up to 30 sec
Delayed up to 1 min
Central
Peripheral
30.0% No. 9 14.3% No. 3 23.5% No. 12
30.0% No. 9 23.8% No. 5 27.4% No. 14
70.0% No. 21 42.8% No. 9 58.8% No. 30
26.6% No. 8 14.3% No. 3 21.5% No. 11
Zone 1 No. 30 Zone 2 No. 21 Total No. 51
Table 2. Vascular–Avascular Junction Findings: Number of Eyes (%)
Abnormal Filling
Methods Fluorescein angiography using RetCam became available for ROP examinations in December 2004 for the neonatal intensive care unit at the Catholic University Hospital in Rome, Italy. From December 2004 to July 2008, 199 inborn preterm infants with gestational age (GA) of ⱕ32 weeks and/or a birth weight (BW) of ⬍1500 g underwent indirect ophthalmoscopy to detect ROP. The mean BW of these infants was 1188 g (range, 350 –2420) and the mean GA was 29.3 weeks (range, 24 –35). Twenty-six infants (13.1%) required treatment, according to early treatment for ROP6 criteria: 22 eyes (11 infants; mean BW, 612.73 g [range, 500 – 740]; mean GA, 25.64 weeks [range, 25–28]) were classified as zone I stage 3 with plus; 8 eyes (4 infants; mean BW, 577.50 g [range, 350 –700]; mean GA, 25.50 weeks [range, 24 –27) were classified as zone I stage 3 without plus; finally 21 eyes (11 infants; mean BW, 770.91 g [range, 500 –1140]; mean GA, 25.91 weeks [range, 24 –29]) were classified as zone II stage 3 with plus. Within a maximum interval of 24 hours after diagnosis, all infants underwent general anesthesia before laser treatment following a previously described protocol.18 Before treatment, RetCam fundus images were obtained and video-digital FA was performed. A 10% solution of fluorescein was intravenously administered as a bolus at a dose of 0.1 ml/kg, followed by an isotonic saline flush. The FA was examined by the treating physicians before conducting laser photocoagulation to provide more detailed information about the status of the eye and to indicate areas of the retina that might be treated. Laser photocoagulation was then performed. Additional digital images were then obtained immediately after laser photocoagulation to determine the adequacy of the retinal ablation.
Zone 1 No. 30 Zone 2 No. 21 Total No. 51
Hyperfluorescent Lesions
Tangles
Naked Shunts
Cotton Wool
Popcorn
Rosary Bead-like
Capillary Tufts
76.6% No. 23 66.7% No. 14 72.5% No. 37
50.0% No. 15 19.0% No. 4 37.3% No. 19
53.3% No. 16 42.9% No. 9 49.0% No. 25
63.3% No. 19 33.3% No. 7 51.0% No. 26
23.3% No. 7 9.5% No. 2 17.6% No. 9
40.0% No. 12 38.1% No. 8 39.2% No. 20
Therefore, a total of 51 sets of digital images including FA were obtained immediately before treatment in eyes with severe ROP. Follow-up digital images were also obtained at 3 and 6 months post-treatment. The local institutional review board authorized data analysis for publication.
Results The FAs were examined to assess whether the morphologic features of the eyes with severe ROP were consistent across eyes or if the features were highly variable. We identified 3 angiographic characteristics of severe ROP: (1) Retinal and choroidal circulation, (2) features of the junction between vascular and avascular retina (V–Av junction) and, (3) abnormalities posterior to the acute phase retinopathy and within the vascularized retina.
Retinal and Choroidal Circulation One of our first observations was the highly variable rate of circulation of the fluorescein dye from the arm to the infant’s eye. The arteriolar phase could appear as late as 53 seconds or as soon as 4 seconds, with a mean value of 12 seconds for the studies reported. A similar variability was observed for the venular phase that could be delayed up to 2 minutes. These findings, summarized in Table 1, seem to confirm the extreme instability blood flow in premature infants with severe ROP. Even if a protocol is standardized,18 we have to emphasize that general anesthesia could play a role in this variability. Although FA is not ideal for examining choroidal circulation, this series of FAs allows some observations. The existing litera-
Figure 1. Fluorescein angiographic image at 41 seconds showing irregular choroidal pattern suggesting areas of hypoperfusion at the posterior pole (A) and at the peripheral retina at 1 minute and 9 seconds (B).
169
Ophthalmology Volume 118, Number 1, January 2011
Figure 2. Fluorescein angiographic images showing junction between vascularized and avascular retina with leakage (white arrows) at 1 minute 42 seconds (A) and persistent leakage at vascularized/avascular junction masking the features of the underlying vessels at 1 minute 13 seconds (B).
Figure 3. Irregular branching of peripheral vessels starting with large arterioles (A) at 35 seconds or small arterioles (B) at 25 seconds, or at the precapillary level (C) at 57seconds.
Figure 4. Color fundus image (A) and fluorescein angiogram (B) at 1 minute 13 seconds in which a new episode of abnormal vascularization is occurring that is peripheral to an area of regressing retinopathy of prematurity.
Figure 5. White plain arrows show an examples of an arteriovenous shunt with no tangle at 1 minute 41 seconds (A) and another example of arteriovenous shunt (arrowheads) with vascular abnormalities beyond the vessels (plain arrows) shown at 2 minutes 2 seconds (B).
170
Lepore et al 䡠 FA in Severe ROP
Figure 6. White plain arrows show perivascular leakage from one vessels near arteriovenous junction at 41 seconds after injection (A); diffuse perivascular leakage (white arrows) posterior to advancing border of vessels at 26 seconds after injection (B); 2 perivascular leakage with “cotton-wool-like” vascular abnormalities (white arrows) shown at 52 seconds after injection (C).
Figure 7. (A) Hyperfluorescent lesions (50 seconds after injection) with well-defined contour (white plain arrows), that could be associated with “popcorn-like” vascular abnormalities on color fundus image (B). Note that these lesions are much less apparent on fundus image, particularly lesions 1, 4, and 5.
Figure 8. Tuft formation in capillary bed (black dotted arrows) shown at 27 seconds after injection (A); focal dilation of capillaries shown by white narrow arrows at 1 minute 7 seconds after injection (B); white arrows indicate rosary-bead-like hyperfluorescent lesions at late venous phase (3 minutes 33 seconds) (C).
171
Ophthalmology Volume 118, Number 1, January 2011 Table 4. Number of Eyes (%) with Macular Abnormalities
Zone 1 No. 30 Zone 2 No. 21 Total No. 51
Figure 9. Excellent image showing many of the changes seen in Figures 6 to 8 but taken at an earlier phase (29 seconds after injection). Note in addition some hemorrhage (black arrows) in the region of the vascularavascular junction, popcorn abnormalities (white plain arrow); focal capillary dilatation (white arrowheads), capillary tuft (white asterisk), rosarybead-like hyperfluorescences (black arrowheads).
ture, although scant, suggests that choroidal circulation is relatively stable during the perinatal period.19 In several of the FAs, we observed an extreme variability of pattern of choroid filling. In eyes with less pigmentation, we noted an irregular choroidal filling, with many persistent hypo/nonperfused areas in the posterior pole surrounded by a hyperfluorescence due to lobular choroidal pattern (Fig 1A); areas of choroidal hypofluorescence sometimes extended to the periphery (Fig 1B). In 1986, Flower19 showed similar “extremely sluggish choroidal flush” in oxygen-exposed kittens and hypothesized that “choroid has the potential to play a major role in maintaining the retina until the retinal vasculature fully develops.” Consistent with his hypothesis, our data show choroid and choriocapillary abnormalities, suggesting the need to examine the role of the choroid in the pathogenesis of functional and morphological damage of the retina, in particular the posterior pole, in the eyes of premature infants with severe ROP.
Features of the Junction between Vascular and Avascular Retina In Table 2, we list a number of findings commonly seen in our series at the V–Av junction. These include the leakage at site of active ROP, a common FA feature when stage 3 ROP is present (Fig 2A). Fluorescein angiography provides excellent visualization
Absence of Foveal Avascular Zone
Hyperfluorescent Lesions
Absence of Wedge
70.0% No. 21 33.3% No. 7 54.9% No. 28
60.0% No. 18 61.9% No. 13 60.8% No. 31
66.7% No. 20 0.0% No. 0 39.2% No. 20
of the transition at V–Av junction with vascular abnormalities beyond the vessels. This is particularly useful in the case of laser treatment of an extremely immature ROP, with “deceptive featureless” V–Av junction in zone I.14 Persistent dye leakage from the ridge sometimes mask the capillary branching features in the late angiography phases (Fig 2B). In addition, FA documents abnormal branching of vessels in eyes with severe ROP. In immature retinas, the normal dichotomous branching is lost.11–14 Several patterns of abnormal branching are observed. We commonly noted an irregular branching starting either at a large arteriolar level (Fig 3A), a small arteriolar level (Fig 3B), or “suddenly” at a precapillary level branching (Fig 3C). This latter feature is also observed in a case when abnormal vascularization occurs beyond areas of regressing ROP (Fig 4) and compared with a fundus photograph. In a less common feature, circumferential vessels, already described in the revisited ICROP,3 runs along the V–Av junction (Fig 5A). They could be interpreted as a “naked” arteriovenous shunt. A similar circumferential vessel is sometimes seen with vascular abnormalities extending beyond the vessels (Fig 5B). Other distinctive features noted in the region of the retina near the V–Av junction include hyperfluorescent lesions. Perivascular dye leakage from the vessels wall is a frequent finding, sometimes seen leaking from an isolated vessel (shown in Fig 6A at 41 seconds) and sometimes leaking from a large number of vessels near the V–Av junction (Fig 6B). Figure 6C shows 2 “cotton-woollike” hyperfluorescent lesions that could be interpreted as part of an exudative process near the V–Av junction. Isolated vascular tufts posterior to the ridge consistent with the description of popcorn in the ICROP20 are sometimes noted (arrows in Fig 7B). Figure 7A shows how these lesions appear in FA: long-lasting hyperfluorescent spots with a well-defined contour. Also, FA documents more “popcorn” lesions than can be seen in the fundus image, perhaps because of exudative processes going in these vascular tufts. Figure 8 shows other findings in the V–Av junction: capillary tuft formations just beyond the V–Av junction, coupled with many other hyperfluorescent areas due to dye leakage (Fig 8A); hyperfluorescent lesions, that is, focal dilatation of capillaries, similar to those observed in diabetic retinopathy (Fig 8B), and rosary-
Table 3. Hypofluorescent Lesions Inside Vascularized Retina: Number of Eyes (%) Hypofluorescent Area
Zone 1 No. 30 Zone 2 No. 21 Total No. 51
172
Loss of Capillary Bed
Periarteriolar Capillary-Free Zone
With Hyperfluorescent Areas
Without Hyperfluorescent Areas
Central
Peripheral
Central
Peripheral
80.0% No. 24 33.3% No. 7 60.8% No. 31
23.3% No. 7 33.3% No. 7 27.5% No. 14
23.3% No. 7 0.0% No. 0 13.7% No. 7
73.3% No. 22 52.4% No. 11 64.7% No. 33
26.7% No. 8 0.0% No. 0 15.7% No. 8
86.7% No. 26 57.1% No. 12 74.5% No. 38
Lepore et al 䡠 FA in Severe ROP
Figure 10. (A) Fluorescein angiogram showing 40 seconds after injection hypofluorescent areas of the retina (white circles). Two of them are surrounded by leakage-associated hyperfluorescence (white asterisk). (B) Fluorescein angiogram at 23 seconds showing aggressive posterior retinopathy of prematurity (ROP) in zone I with massive extent of hypofluorescent areas. (C) Fluorescein angiogram of an eye with severe ROP in zone II showing extensive capillary nonperfusion at 35 seconds after injection.
bead-like hyperfluorescent lesions inside the vessels (Fig 8C). Some eyes show all of these features on FA at the same time, as shown by Figure 9, in addition with retinal hemorrhages from extraretinal neovascular proliferation.
Findings within the Vascularized Retina and at the Posterior Pole These findings illustrate features of arrested vascular development at the edge between vascular and avascular retina, where the vascular abnormality is well known to show the maximum effect. However, FA in our case series shows vascular abnormalities occurring well inside the edge of the vascularized retina (Table 3). Some of the abnormalities that we observe are already reported in smaller series,16,17 and some are new. In particular, the FA case series in this report demonstrate that posterior retina is also heavily affected by ROP. Rather than being immune to damage, the macula is involved also in many zone 2 ROP cases (Table 4). We observed areas of hypofluorescence, which were not observed on ophthalmoscopy; some of these areas are surrounded by hyperfluorescent spots owing to persistent dye leakage (Fig 10A). Hypofluorescence owing to retinal or choroidal vascular filling defects at the posterior pole is a common finding in the aggressive posterior ROP, where the macula region pole is heavily involved (Fig 10B). Sometimes hypofluorescent areas are also observed in
severe zone II ROP (Fig 10C). A massive loss of retinal capillary bed and/or lack of choroidal perfusion could explain this impressive feature. In addition to diffuse areas of hypoperfusion, periarteriolar loss of capillary bed is a consistent finding in this case series of FA. Figure 11A shows an extensive, nonperfused region of the retina with an evident loss of capillaries around major vessels. In general, periarteriolar capillary-free zone is observed limited to peripheral retina, although it occasionally extends from the posterior pole to the optic disk (Fig 11B). As far as we know, this is the first report of periarteriolar capillary-free zone in premature retina. Using a kitten model of oxygen-induced retinopathy, Michaelson et al21 showed that oxygen can influence the extension of the periarteriolar capillary free zone, thus playing a role in capillary remodelling.22–24 Further studies are needed to clarify the value of this observation in the human ROP and in particular whether oxygen causes the obliteration of the preexisting vessels or prevents their growth. In this FA case series, various macular abnormalities are noted (Table 4). Absence of a foveal avascular zone is a frequent observation in our series (Fig 12A). This finding is consistent with report of Yokoi et al16 and Mintz-Hittner et al.25 The latter in particular showed that the fine meshwork of inner retinal vessels observed during vasculogenesis never fully regresses in preterm infants ⱕ30 weeks GA at birth, suggesting that small or absent foveal avascular zone may remain as an historic mark of prematurity.
Figure 11. (A) Fluorescein angiogram at 33 seconds shows extensive capillary nonperfusion areas (white circles), hyperfluorescence associated with leakage (white arrows), and capillary free zone around arterioles (black arrowhead) that are largely peripheral. (B) Capillary-free zone around arterioles at 34 seconds after injection extending posteriorly to the optic disc (black narrows arrows).
173
Ophthalmology Volume 118, Number 1, January 2011
Figure 12. (A) Absence of foveal avascular zone seen at 32 seconds in eyes with aggressive posterior retinopathy of prematurity. (B) Fluorescein angiogram with diffuse dye leakage at 22 seconds in the region of the macula (plain white circle), suggesting an exudation.
Because of the absence of the foveal avascular zone, even with FA, it is often impossible to exactly locate the macula. Because the location of the macula is the basis for defining the retinal areas of zones 1, 2, and 3, difficulties in macula recognition could help to explain the variability among different clinical experiences in the incidence of zone 1 ROP. It also suggests another “revisiting” of ICROP would be useful. We also noted other macular abnormalities including diffuse hypoperfusion (Fig 10B) or, more significantly, hyperfluorescent lesions (Fig 12B), suggesting exudation beneath the macula. Finally, we note the absence of the wedge (Fig 13), described as one of the characteristics that differentiate zone I versus zone II ROP,10 is sometimes lost in the very immature eye (Fig 10B). This finding is consistent with the hypothesis that a zone 1/zone 2 hybrid form of ROP is more frequently observed than typical zone 1 ROP presentation. Advances in imaging are transforming our understanding of angiogenesis-related pathogenesis of ROP. Techniques such as the FA could play a central role in understanding mechanical and functional implications of angiogenesis in ROP. By using FA, we have highlighted vessels as small as 25 microns and assessed neovascularization, vessel leakage, loss of capillary beds, and other hypoperfused areas of the retina and choroid that could represent another driving force of vascular endothelial growth factor–induced clinical signs of ROP. The question remains how these findings will help to
Figure 13. circumferential appearance of the vascular abnormality. No “wedge” is observed.
174
explain not only the retinal vascular development but also the neuroretinal migration and differentiation. Answering that question will open new therapeutic options and perspectives in ROP.
References 1. Committee for the Classification of Retinopathy of Prematurity. An international classification of retinopathy of prematurity. Arch Ophthalmol 1984;102:1130 – 4. 2. International Committee for the Classification of the Late Stages of Retinopathy of Prematurity. An international classification of retinopathy of prematurity. II. The classification of retinal detachment. Arch Ophthalmol 1987;105:906 –12. 3. International Committee for the Classification of Retinopathy of Prematurity. The International Classification of Retinopathy of Prematurity revisited. Arch Ophthalmol 2005;123: 991–9. 4. Cryotherapy for Retinopathy of Prematurity Cooperative Group. Multicenter trial of cryotherapy for retinopathy: preliminary results. Arch Ophthalmol 1988;105:471–9. 5. Cryotherapy for Retinopathy of Prematurity Cooperative Group. Multicenter trial of cryotherapy for retinopathy: ophthalmological outcomes at 10 years. Arch Ophthalmol 2001; 119:1110 – 8. 6. 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 –96. 7. Reynolds JD, Dobson V, Quinn GE, et al, CRYO-ROP Cooperative Group, LIGHT-ROP Cooperative Group. Evidencebased screening for retinopathy of prematurity: natural history data from CRYO-ROP and LIGHT-ROP studies. Arch Ophthalmol 2002;120:1470 – 6. 8. Shih SC, Ju M, Liu N, Smith LE. Selective stimulation of VEGFR-1 prevents oxygen-induced retinal vascular degeneration in retinopathy of prematurity. J Clin Invest 2003;112: 50 –7. 9. Hellstrom A, Peruzzi C, Ju M, et al. Low IGF-I suppresses VEGF-survival signaling in retinal endothelial cells: direct correlation with clinical retinopathy of prematurity. Proc Natl Acad Sci U S A 2001;98:5804 – 8. 10. Flynn JT, Chan-Ling T. Retinopathy of prematurity: two distinct mechanism that underlie zone 1 and zone 2 disease. Am J Ophthalmol 2006;142:46 –59.
Lepore et al 䡠 FA in Severe ROP 11. Cantolino SJ, O’Grady GE, Herrera JA, et al. Ophthalmoscopic monitoring of oxygen therapy in premature infants: fluorescein angiography in acute retrolental fibroplasia. Am J Ophthalmol 1971;72:322–31. 12. Flynn JT, O’Grady GE, Herrera J, et al. Retrolental fibroplasia: I. Clinical observations. Arch Ophthalmol 1977; 95:217–23. 13. Flynn JT, Cassady J, Essner D, et al. Fluorescein angiography in retrolental fibroplasia: experience from 1969 –1977. Ophthalmology 1979;86:1700 –23. 14. Schulenburg WE, Tsanaktsidis G. Variations in the morphology of retinopathy of prematurity in extremely low birthweight infants. Br J Ophthalmol 2004;88:1500 –3. 15. Ng EY, Lanigan B, O’Keefe M. Fundus fluorescein angiography in the screening for and management of retinopathy of prematurity. J Pediatric Ophthalmol Strabismus 2006;43: 85–90. 16. Yokoi T, Hiraoka M, Miyamoto M, et al. Vascular abnormalities in aggressive posterior retinopathy of prematurity detected by fluorescein angiography. Ophthalmology 2009;116: 1377– 82. 17. Azad R, Chandra P, Khan MA, Darswal A. Role of intravenous fluorescein angiography in early detection and regression of retinopathy of prematurity. J Pediatric Ophthalmol Strabismus 2008;45:36 –9. 18. Sammartino M, Bocci MG, Ferro G, et al. Efficacy and safety of continuous intravenous infusion of remifentanil in preterm
19.
20.
21.
22.
23.
24.
25.
infants undergoing laser therapy in retinopathy of prematurity: clinical experience. Paediatr Anaesth 2003;13:596 – 602. Flower RW. Perinatal retinal vascular physiology. In: Silverman WA, Flynn JT, eds. Retinopathy of Prematurity. Boston: Blackwell Sci. Publ.; 1985:97–129. Contemporary Issues in Fetal and Neonatal Medicine. vol. 2. Quinn G. Retinopathy of prematurity: natural history and classification. In: Flynn JT, Tasman W, eds. Retinopathy of Prematurity: A Clinician’s Guide. New York: SpringerVerlag; 1992:7–22. Michaelson IC, Benezra D, Berson D. Possible metabolic mechanism modulating blood vessel development in the inner eye and their significance for vascular pathology in the definitive eye. Metab Pediatr Syst Ophthalmol 1982;6:1–10. Phelps DL. Oxygen and developmental retinal capillary remodelling in the kitten. Invest Ophthalmol Vis Sci 1990;31: 2194 –200. Chan-Ling T, Tout S, Holländer H, Stone J. Vascular changes and their mechanisms in the feline model of retinopathy of prematurity. Invest Ophthalmol Vis Sci 1992;33:2128 – 47. McLeod DS, D’Anna SA, Lutty GA. Clinical and histopathologic features of canine oxygen-induced proliferative retinopathy. Invest Ophthalmol Vis Sci 1998;39:1918 –32. Mintz-Hittner HA, Knight-Nanan DM, Satriano DR, Kretzer FL. A small foveal avascular zone may be an historic mark of prematurity. Ophthalmology 1999;106:1409 –13.
Footnotes and Financial Disclosures Originally received: October 21, 2009. Final revision: April 9, 2010. Accepted: April 13, 2010. Available online: August 14, 2010.
Financial Disclosure(s): The authors have no proprietary or commercial interest in any of the materials discussed in this article. Manuscript no. 2009-1472.
1
Department Ophthalmology and Dept Anesthesiology, Catholic University of Sacred Heart, Rome, Italy. 2
Children’s Hospital of Philadelphia University of Pennsylvania, Philadelphia, Pennsylvania.
Correspondence: Domenico Lepore, MD, Dept Ophthalmology, Catholic University of Sacred Heart, Largo Francesco Vito 8, 00168 Rome, Italy.
175
Retinopathy of prematurity (ROP) is a disease of developing blood vessels in the retina of the premature infant. In 1984, an International Committee of ROP experts developed a classification (International Classification for ROP [ICROP]) characterized by the location, stage, and extent of the disease based on the ophthalmoscopic findings.1,2 In its revisited form,3 ICROP has been the common language used not only to manage ROP clinically,4 –7 but also to provide a framework for studying basic mechanisms.8,9 Based on ophthalmoscopic appearance of ROP in zones 1 and 2, Flynn and Chan-Lin10 hypothesized 2 distinct mechanisms in the pathogenesis of ROP: In particular, the authors suggested a correlation between zone I ROP features and vasculogenic process in the posterior retina. To better illustrate their hypothesis, the authors cited previous fluorescein angiographic (FA) studies of ROP.11–13 These pioneering studies clearly showed that FA, although technically challenging, could represent a major tool to study the mechanism of vessel formation and maturation in the pre-
168
© 2011 by the American Academy of Ophthalmology Published by Elsevier Inc.
mature retina. Schulenburg and Tsanaktsids14 in 2004 used FA to investigate vascular morphology in 6 cases of extremely low– birth-weight ROP. This study showed a different ROP morphology without classic shunt formation and a poorly developed capillary bed in already vascularized retina. In the late 1990s with the advent of digital imaging of the eyes of infants at risk for ROP, FA became easier and safer to perform in the neonatal intensive care unit using such instruments as the RetCam (Clarity, Pleasanton, CA). In 2006, Ng et al15 demonstrated that clear angiograms can be obtained as part of ROP screening. In the same year, we also presented preliminary results of FA in zone I ROP (Pagliara MM, Lepore D, Baldascino A, et al. Fluorescein angiographic findings in zone I ROP [abstract]. Invest Ophthalmol Vis Sci 2006;47:E-Abstract 5302). Since then, other groups have reported small series of FA findings in ROP.16,17 Building on these studies, we have conducted FA in eyes with severe ROP that are undergoing laser photoablation to examine systematically the vascular abnormalities of such eyes. ISSN 0161-6420/11/$–see front matter doi:10.1016/j.ophtha.2010.04.021
Lepore et al 䡠 FA in Severe ROP Table 1. Circulation Patterns: Number of Eyes (%) in our Series Showing Arterial Phase Appearance Delayed up to 30 Seconds, Venous Phase Appearance Delayed up to 1 Minute, and Abnormal Choroidal Filling (as described in Fig 1A, B)
Branching
Choroid
Arterial Phase
Venous Phase
Delayed up to 30 sec
Delayed up to 1 min
Central
Peripheral
30.0% No. 9 14.3% No. 3 23.5% No. 12
30.0% No. 9 23.8% No. 5 27.4% No. 14
70.0% No. 21 42.8% No. 9 58.8% No. 30
26.6% No. 8 14.3% No. 3 21.5% No. 11
Zone 1 No. 30 Zone 2 No. 21 Total No. 51
Table 2. Vascular–Avascular Junction Findings: Number of Eyes (%)
Abnormal Filling
Methods Fluorescein angiography using RetCam became available for ROP examinations in December 2004 for the neonatal intensive care unit at the Catholic University Hospital in Rome, Italy. From December 2004 to July 2008, 199 inborn preterm infants with gestational age (GA) of ⱕ32 weeks and/or a birth weight (BW) of ⬍1500 g underwent indirect ophthalmoscopy to detect ROP. The mean BW of these infants was 1188 g (range, 350 –2420) and the mean GA was 29.3 weeks (range, 24 –35). Twenty-six infants (13.1%) required treatment, according to early treatment for ROP6 criteria: 22 eyes (11 infants; mean BW, 612.73 g [range, 500 – 740]; mean GA, 25.64 weeks [range, 25–28]) were classified as zone I stage 3 with plus; 8 eyes (4 infants; mean BW, 577.50 g [range, 350 –700]; mean GA, 25.50 weeks [range, 24 –27) were classified as zone I stage 3 without plus; finally 21 eyes (11 infants; mean BW, 770.91 g [range, 500 –1140]; mean GA, 25.91 weeks [range, 24 –29]) were classified as zone II stage 3 with plus. Within a maximum interval of 24 hours after diagnosis, all infants underwent general anesthesia before laser treatment following a previously described protocol.18 Before treatment, RetCam fundus images were obtained and video-digital FA was performed. A 10% solution of fluorescein was intravenously administered as a bolus at a dose of 0.1 ml/kg, followed by an isotonic saline flush. The FA was examined by the treating physicians before conducting laser photocoagulation to provide more detailed information about the status of the eye and to indicate areas of the retina that might be treated. Laser photocoagulation was then performed. Additional digital images were then obtained immediately after laser photocoagulation to determine the adequacy of the retinal ablation.
Zone 1 No. 30 Zone 2 No. 21 Total No. 51
Hyperfluorescent Lesions
Tangles
Naked Shunts
Cotton Wool
Popcorn
Rosary Bead-like
Capillary Tufts
76.6% No. 23 66.7% No. 14 72.5% No. 37
50.0% No. 15 19.0% No. 4 37.3% No. 19
53.3% No. 16 42.9% No. 9 49.0% No. 25
63.3% No. 19 33.3% No. 7 51.0% No. 26
23.3% No. 7 9.5% No. 2 17.6% No. 9
40.0% No. 12 38.1% No. 8 39.2% No. 20
Therefore, a total of 51 sets of digital images including FA were obtained immediately before treatment in eyes with severe ROP. Follow-up digital images were also obtained at 3 and 6 months post-treatment. The local institutional review board authorized data analysis for publication.
Results The FAs were examined to assess whether the morphologic features of the eyes with severe ROP were consistent across eyes or if the features were highly variable. We identified 3 angiographic characteristics of severe ROP: (1) Retinal and choroidal circulation, (2) features of the junction between vascular and avascular retina (V–Av junction) and, (3) abnormalities posterior to the acute phase retinopathy and within the vascularized retina.
Retinal and Choroidal Circulation One of our first observations was the highly variable rate of circulation of the fluorescein dye from the arm to the infant’s eye. The arteriolar phase could appear as late as 53 seconds or as soon as 4 seconds, with a mean value of 12 seconds for the studies reported. A similar variability was observed for the venular phase that could be delayed up to 2 minutes. These findings, summarized in Table 1, seem to confirm the extreme instability blood flow in premature infants with severe ROP. Even if a protocol is standardized,18 we have to emphasize that general anesthesia could play a role in this variability. Although FA is not ideal for examining choroidal circulation, this series of FAs allows some observations. The existing litera-
Figure 1. Fluorescein angiographic image at 41 seconds showing irregular choroidal pattern suggesting areas of hypoperfusion at the posterior pole (A) and at the peripheral retina at 1 minute and 9 seconds (B).
169
Ophthalmology Volume 118, Number 1, January 2011
Figure 2. Fluorescein angiographic images showing junction between vascularized and avascular retina with leakage (white arrows) at 1 minute 42 seconds (A) and persistent leakage at vascularized/avascular junction masking the features of the underlying vessels at 1 minute 13 seconds (B).
Figure 3. Irregular branching of peripheral vessels starting with large arterioles (A) at 35 seconds or small arterioles (B) at 25 seconds, or at the precapillary level (C) at 57seconds.
Figure 4. Color fundus image (A) and fluorescein angiogram (B) at 1 minute 13 seconds in which a new episode of abnormal vascularization is occurring that is peripheral to an area of regressing retinopathy of prematurity.
Figure 5. White plain arrows show an examples of an arteriovenous shunt with no tangle at 1 minute 41 seconds (A) and another example of arteriovenous shunt (arrowheads) with vascular abnormalities beyond the vessels (plain arrows) shown at 2 minutes 2 seconds (B).
170
Lepore et al 䡠 FA in Severe ROP
Figure 6. White plain arrows show perivascular leakage from one vessels near arteriovenous junction at 41 seconds after injection (A); diffuse perivascular leakage (white arrows) posterior to advancing border of vessels at 26 seconds after injection (B); 2 perivascular leakage with “cotton-wool-like” vascular abnormalities (white arrows) shown at 52 seconds after injection (C).
Figure 7. (A) Hyperfluorescent lesions (50 seconds after injection) with well-defined contour (white plain arrows), that could be associated with “popcorn-like” vascular abnormalities on color fundus image (B). Note that these lesions are much less apparent on fundus image, particularly lesions 1, 4, and 5.
Figure 8. Tuft formation in capillary bed (black dotted arrows) shown at 27 seconds after injection (A); focal dilation of capillaries shown by white narrow arrows at 1 minute 7 seconds after injection (B); white arrows indicate rosary-bead-like hyperfluorescent lesions at late venous phase (3 minutes 33 seconds) (C).
171
Ophthalmology Volume 118, Number 1, January 2011 Table 4. Number of Eyes (%) with Macular Abnormalities
Zone 1 No. 30 Zone 2 No. 21 Total No. 51
Figure 9. Excellent image showing many of the changes seen in Figures 6 to 8 but taken at an earlier phase (29 seconds after injection). Note in addition some hemorrhage (black arrows) in the region of the vascularavascular junction, popcorn abnormalities (white plain arrow); focal capillary dilatation (white arrowheads), capillary tuft (white asterisk), rosarybead-like hyperfluorescences (black arrowheads).
ture, although scant, suggests that choroidal circulation is relatively stable during the perinatal period.19 In several of the FAs, we observed an extreme variability of pattern of choroid filling. In eyes with less pigmentation, we noted an irregular choroidal filling, with many persistent hypo/nonperfused areas in the posterior pole surrounded by a hyperfluorescence due to lobular choroidal pattern (Fig 1A); areas of choroidal hypofluorescence sometimes extended to the periphery (Fig 1B). In 1986, Flower19 showed similar “extremely sluggish choroidal flush” in oxygen-exposed kittens and hypothesized that “choroid has the potential to play a major role in maintaining the retina until the retinal vasculature fully develops.” Consistent with his hypothesis, our data show choroid and choriocapillary abnormalities, suggesting the need to examine the role of the choroid in the pathogenesis of functional and morphological damage of the retina, in particular the posterior pole, in the eyes of premature infants with severe ROP.
Features of the Junction between Vascular and Avascular Retina In Table 2, we list a number of findings commonly seen in our series at the V–Av junction. These include the leakage at site of active ROP, a common FA feature when stage 3 ROP is present (Fig 2A). Fluorescein angiography provides excellent visualization
Absence of Foveal Avascular Zone
Hyperfluorescent Lesions
Absence of Wedge
70.0% No. 21 33.3% No. 7 54.9% No. 28
60.0% No. 18 61.9% No. 13 60.8% No. 31
66.7% No. 20 0.0% No. 0 39.2% No. 20
of the transition at V–Av junction with vascular abnormalities beyond the vessels. This is particularly useful in the case of laser treatment of an extremely immature ROP, with “deceptive featureless” V–Av junction in zone I.14 Persistent dye leakage from the ridge sometimes mask the capillary branching features in the late angiography phases (Fig 2B). In addition, FA documents abnormal branching of vessels in eyes with severe ROP. In immature retinas, the normal dichotomous branching is lost.11–14 Several patterns of abnormal branching are observed. We commonly noted an irregular branching starting either at a large arteriolar level (Fig 3A), a small arteriolar level (Fig 3B), or “suddenly” at a precapillary level branching (Fig 3C). This latter feature is also observed in a case when abnormal vascularization occurs beyond areas of regressing ROP (Fig 4) and compared with a fundus photograph. In a less common feature, circumferential vessels, already described in the revisited ICROP,3 runs along the V–Av junction (Fig 5A). They could be interpreted as a “naked” arteriovenous shunt. A similar circumferential vessel is sometimes seen with vascular abnormalities extending beyond the vessels (Fig 5B). Other distinctive features noted in the region of the retina near the V–Av junction include hyperfluorescent lesions. Perivascular dye leakage from the vessels wall is a frequent finding, sometimes seen leaking from an isolated vessel (shown in Fig 6A at 41 seconds) and sometimes leaking from a large number of vessels near the V–Av junction (Fig 6B). Figure 6C shows 2 “cotton-woollike” hyperfluorescent lesions that could be interpreted as part of an exudative process near the V–Av junction. Isolated vascular tufts posterior to the ridge consistent with the description of popcorn in the ICROP20 are sometimes noted (arrows in Fig 7B). Figure 7A shows how these lesions appear in FA: long-lasting hyperfluorescent spots with a well-defined contour. Also, FA documents more “popcorn” lesions than can be seen in the fundus image, perhaps because of exudative processes going in these vascular tufts. Figure 8 shows other findings in the V–Av junction: capillary tuft formations just beyond the V–Av junction, coupled with many other hyperfluorescent areas due to dye leakage (Fig 8A); hyperfluorescent lesions, that is, focal dilatation of capillaries, similar to those observed in diabetic retinopathy (Fig 8B), and rosary-
Table 3. Hypofluorescent Lesions Inside Vascularized Retina: Number of Eyes (%) Hypofluorescent Area
Zone 1 No. 30 Zone 2 No. 21 Total No. 51
172
Loss of Capillary Bed
Periarteriolar Capillary-Free Zone
With Hyperfluorescent Areas
Without Hyperfluorescent Areas
Central
Peripheral
Central
Peripheral
80.0% No. 24 33.3% No. 7 60.8% No. 31
23.3% No. 7 33.3% No. 7 27.5% No. 14
23.3% No. 7 0.0% No. 0 13.7% No. 7
73.3% No. 22 52.4% No. 11 64.7% No. 33
26.7% No. 8 0.0% No. 0 15.7% No. 8
86.7% No. 26 57.1% No. 12 74.5% No. 38
Lepore et al 䡠 FA in Severe ROP
Figure 10. (A) Fluorescein angiogram showing 40 seconds after injection hypofluorescent areas of the retina (white circles). Two of them are surrounded by leakage-associated hyperfluorescence (white asterisk). (B) Fluorescein angiogram at 23 seconds showing aggressive posterior retinopathy of prematurity (ROP) in zone I with massive extent of hypofluorescent areas. (C) Fluorescein angiogram of an eye with severe ROP in zone II showing extensive capillary nonperfusion at 35 seconds after injection.
bead-like hyperfluorescent lesions inside the vessels (Fig 8C). Some eyes show all of these features on FA at the same time, as shown by Figure 9, in addition with retinal hemorrhages from extraretinal neovascular proliferation.
Findings within the Vascularized Retina and at the Posterior Pole These findings illustrate features of arrested vascular development at the edge between vascular and avascular retina, where the vascular abnormality is well known to show the maximum effect. However, FA in our case series shows vascular abnormalities occurring well inside the edge of the vascularized retina (Table 3). Some of the abnormalities that we observe are already reported in smaller series,16,17 and some are new. In particular, the FA case series in this report demonstrate that posterior retina is also heavily affected by ROP. Rather than being immune to damage, the macula is involved also in many zone 2 ROP cases (Table 4). We observed areas of hypofluorescence, which were not observed on ophthalmoscopy; some of these areas are surrounded by hyperfluorescent spots owing to persistent dye leakage (Fig 10A). Hypofluorescence owing to retinal or choroidal vascular filling defects at the posterior pole is a common finding in the aggressive posterior ROP, where the macula region pole is heavily involved (Fig 10B). Sometimes hypofluorescent areas are also observed in
severe zone II ROP (Fig 10C). A massive loss of retinal capillary bed and/or lack of choroidal perfusion could explain this impressive feature. In addition to diffuse areas of hypoperfusion, periarteriolar loss of capillary bed is a consistent finding in this case series of FA. Figure 11A shows an extensive, nonperfused region of the retina with an evident loss of capillaries around major vessels. In general, periarteriolar capillary-free zone is observed limited to peripheral retina, although it occasionally extends from the posterior pole to the optic disk (Fig 11B). As far as we know, this is the first report of periarteriolar capillary-free zone in premature retina. Using a kitten model of oxygen-induced retinopathy, Michaelson et al21 showed that oxygen can influence the extension of the periarteriolar capillary free zone, thus playing a role in capillary remodelling.22–24 Further studies are needed to clarify the value of this observation in the human ROP and in particular whether oxygen causes the obliteration of the preexisting vessels or prevents their growth. In this FA case series, various macular abnormalities are noted (Table 4). Absence of a foveal avascular zone is a frequent observation in our series (Fig 12A). This finding is consistent with report of Yokoi et al16 and Mintz-Hittner et al.25 The latter in particular showed that the fine meshwork of inner retinal vessels observed during vasculogenesis never fully regresses in preterm infants ⱕ30 weeks GA at birth, suggesting that small or absent foveal avascular zone may remain as an historic mark of prematurity.
Figure 11. (A) Fluorescein angiogram at 33 seconds shows extensive capillary nonperfusion areas (white circles), hyperfluorescence associated with leakage (white arrows), and capillary free zone around arterioles (black arrowhead) that are largely peripheral. (B) Capillary-free zone around arterioles at 34 seconds after injection extending posteriorly to the optic disc (black narrows arrows).
173
Ophthalmology Volume 118, Number 1, January 2011
Figure 12. (A) Absence of foveal avascular zone seen at 32 seconds in eyes with aggressive posterior retinopathy of prematurity. (B) Fluorescein angiogram with diffuse dye leakage at 22 seconds in the region of the macula (plain white circle), suggesting an exudation.
Because of the absence of the foveal avascular zone, even with FA, it is often impossible to exactly locate the macula. Because the location of the macula is the basis for defining the retinal areas of zones 1, 2, and 3, difficulties in macula recognition could help to explain the variability among different clinical experiences in the incidence of zone 1 ROP. It also suggests another “revisiting” of ICROP would be useful. We also noted other macular abnormalities including diffuse hypoperfusion (Fig 10B) or, more significantly, hyperfluorescent lesions (Fig 12B), suggesting exudation beneath the macula. Finally, we note the absence of the wedge (Fig 13), described as one of the characteristics that differentiate zone I versus zone II ROP,10 is sometimes lost in the very immature eye (Fig 10B). This finding is consistent with the hypothesis that a zone 1/zone 2 hybrid form of ROP is more frequently observed than typical zone 1 ROP presentation. Advances in imaging are transforming our understanding of angiogenesis-related pathogenesis of ROP. Techniques such as the FA could play a central role in understanding mechanical and functional implications of angiogenesis in ROP. By using FA, we have highlighted vessels as small as 25 microns and assessed neovascularization, vessel leakage, loss of capillary beds, and other hypoperfused areas of the retina and choroid that could represent another driving force of vascular endothelial growth factor–induced clinical signs of ROP. The question remains how these findings will help to
Figure 13. circumferential appearance of the vascular abnormality. No “wedge” is observed.
174
explain not only the retinal vascular development but also the neuroretinal migration and differentiation. Answering that question will open new therapeutic options and perspectives in ROP.
References 1. Committee for the Classification of Retinopathy of Prematurity. An international classification of retinopathy of prematurity. Arch Ophthalmol 1984;102:1130 – 4. 2. International Committee for the Classification of the Late Stages of Retinopathy of Prematurity. An international classification of retinopathy of prematurity. II. The classification of retinal detachment. Arch Ophthalmol 1987;105:906 –12. 3. International Committee for the Classification of Retinopathy of Prematurity. The International Classification of Retinopathy of Prematurity revisited. Arch Ophthalmol 2005;123: 991–9. 4. Cryotherapy for Retinopathy of Prematurity Cooperative Group. Multicenter trial of cryotherapy for retinopathy: preliminary results. Arch Ophthalmol 1988;105:471–9. 5. Cryotherapy for Retinopathy of Prematurity Cooperative Group. Multicenter trial of cryotherapy for retinopathy: ophthalmological outcomes at 10 years. Arch Ophthalmol 2001; 119:1110 – 8. 6. 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 –96. 7. Reynolds JD, Dobson V, Quinn GE, et al, CRYO-ROP Cooperative Group, LIGHT-ROP Cooperative Group. Evidencebased screening for retinopathy of prematurity: natural history data from CRYO-ROP and LIGHT-ROP studies. Arch Ophthalmol 2002;120:1470 – 6. 8. Shih SC, Ju M, Liu N, Smith LE. Selective stimulation of VEGFR-1 prevents oxygen-induced retinal vascular degeneration in retinopathy of prematurity. J Clin Invest 2003;112: 50 –7. 9. Hellstrom A, Peruzzi C, Ju M, et al. Low IGF-I suppresses VEGF-survival signaling in retinal endothelial cells: direct correlation with clinical retinopathy of prematurity. Proc Natl Acad Sci U S A 2001;98:5804 – 8. 10. Flynn JT, Chan-Ling T. Retinopathy of prematurity: two distinct mechanism that underlie zone 1 and zone 2 disease. Am J Ophthalmol 2006;142:46 –59.
Lepore et al 䡠 FA in Severe ROP 11. Cantolino SJ, O’Grady GE, Herrera JA, et al. Ophthalmoscopic monitoring of oxygen therapy in premature infants: fluorescein angiography in acute retrolental fibroplasia. Am J Ophthalmol 1971;72:322–31. 12. Flynn JT, O’Grady GE, Herrera J, et al. Retrolental fibroplasia: I. Clinical observations. Arch Ophthalmol 1977; 95:217–23. 13. Flynn JT, Cassady J, Essner D, et al. Fluorescein angiography in retrolental fibroplasia: experience from 1969 –1977. Ophthalmology 1979;86:1700 –23. 14. Schulenburg WE, Tsanaktsidis G. Variations in the morphology of retinopathy of prematurity in extremely low birthweight infants. Br J Ophthalmol 2004;88:1500 –3. 15. Ng EY, Lanigan B, O’Keefe M. Fundus fluorescein angiography in the screening for and management of retinopathy of prematurity. J Pediatric Ophthalmol Strabismus 2006;43: 85–90. 16. Yokoi T, Hiraoka M, Miyamoto M, et al. Vascular abnormalities in aggressive posterior retinopathy of prematurity detected by fluorescein angiography. Ophthalmology 2009;116: 1377– 82. 17. Azad R, Chandra P, Khan MA, Darswal A. Role of intravenous fluorescein angiography in early detection and regression of retinopathy of prematurity. J Pediatric Ophthalmol Strabismus 2008;45:36 –9. 18. Sammartino M, Bocci MG, Ferro G, et al. Efficacy and safety of continuous intravenous infusion of remifentanil in preterm
19.
20.
21.
22.
23.
24.
25.
infants undergoing laser therapy in retinopathy of prematurity: clinical experience. Paediatr Anaesth 2003;13:596 – 602. Flower RW. Perinatal retinal vascular physiology. In: Silverman WA, Flynn JT, eds. Retinopathy of Prematurity. Boston: Blackwell Sci. Publ.; 1985:97–129. Contemporary Issues in Fetal and Neonatal Medicine. vol. 2. Quinn G. Retinopathy of prematurity: natural history and classification. In: Flynn JT, Tasman W, eds. Retinopathy of Prematurity: A Clinician’s Guide. New York: SpringerVerlag; 1992:7–22. Michaelson IC, Benezra D, Berson D. Possible metabolic mechanism modulating blood vessel development in the inner eye and their significance for vascular pathology in the definitive eye. Metab Pediatr Syst Ophthalmol 1982;6:1–10. Phelps DL. Oxygen and developmental retinal capillary remodelling in the kitten. Invest Ophthalmol Vis Sci 1990;31: 2194 –200. Chan-Ling T, Tout S, Holländer H, Stone J. Vascular changes and their mechanisms in the feline model of retinopathy of prematurity. Invest Ophthalmol Vis Sci 1992;33:2128 – 47. McLeod DS, D’Anna SA, Lutty GA. Clinical and histopathologic features of canine oxygen-induced proliferative retinopathy. Invest Ophthalmol Vis Sci 1998;39:1918 –32. Mintz-Hittner HA, Knight-Nanan DM, Satriano DR, Kretzer FL. A small foveal avascular zone may be an historic mark of prematurity. Ophthalmology 1999;106:1409 –13.
Footnotes and Financial Disclosures Originally received: October 21, 2009. Final revision: April 9, 2010. Accepted: April 13, 2010. Available online: August 14, 2010.
Financial Disclosure(s): The authors have no proprietary or commercial interest in any of the materials discussed in this article. Manuscript no. 2009-1472.
1
Department Ophthalmology and Dept Anesthesiology, Catholic University of Sacred Heart, Rome, Italy. 2
Children’s Hospital of Philadelphia University of Pennsylvania, Philadelphia, Pennsylvania.
Correspondence: Domenico Lepore, MD, Dept Ophthalmology, Catholic University of Sacred Heart, Largo Francesco Vito 8, 00168 Rome, Italy.
175