- Email: [email protected]
A small foveal avascular zone may be an historic mark of prematurity
A Small Foveal Avascular Zone May Be an Historic Mark of Prematurity Helen A. Mintz–Hittner, MD,1,2 Donna M. Knight–Nanan, MD,1 Dale R. Satriano, Frank L. Kretzer, PhD2 Objective: To compare in children the area and diameter of the foveal avascular zone (FAZ) of former preterm infants, when no significant retinopathy of prematurity (ROP) developed, to the area and diameter of the FAZ of former term infants. Design: Retrospective observational case series and literature review. Participants: Forty-nine children (39 former preterm infants and 10 former term infants) between the ages of 1 and 17 years had fluorescein angiograms. All of these children had been appropriate weight for gestational age at birth and had no genetic disorders. Neither eye of any of these children had any macular ectopia or vessel traction, had been treated for active ROP, had developed active ROP ⬎stage 3 mild, or had any refractive error ⬎⫾ five diopters. Every child had a visual acuity of 20/40 or better in both eyes. Methods: The area and greatest diameter of the FAZ were measured using digital image analysis of masked fundus fluorescein angiograms. Variables of gender, race, multiple birth, gestational age, birth weight, ROP stage, age, and refraction at the time of fluorescein angiography, and final visual acuity were recorded. Results: Increasing FAZ area and greatest diameter correlated significantly with increasing gestational age and birth weight: FAZ area (m2) versus gestational age (weeks) (R/F/P ⫽ 0.88/166.70/⬍0.0001); FAZ greatest diameter (m) versus gestational age (weeks) (R/F/P ⫽ 0.87/151.10/⬍0.0001); FAZ area (m2) versus birth weight (g) (R/F/P ⫽ 0.88/167.06/⬍0.0001); and FAZ greatest diameter (m) versus birth weight (g) (R/F/P ⫽ 0.87/148.74/ ⬍0.0001). A small or absent FAZ was found in all former preterm infants who had been ⱕ30 weeks gestational age or had weighed ⱕ1100 g at birth. A normal FAZ was present in all children who had been ⱖ36 weeks gestational age or had weighed ⱖ2650 g at birth. None of the other parameters studied correlated with FAZ area or greatest diameter. Conclusion: This study provides evidence that the FAZ in developing humans is initially densely vascularized with a fine meshwork of inner retinal vessels during vasculogenesis. This vascular meshwork undergoes regression by apoptosis in all infants ⱖ36 weeks gestational age at birth to form a normal FAZ, but apoptosis almost never occurs in preterm infants ⱕ30 weeks gestational age at birth. Although there is no effect on final visual acuity, a small or absent FAZ may be an historic mark of prematurity. Ophthalmology 1999;106:1409 –1413 The adult human fovea is specialized for color vision and better visual acuity than other areas of the retina. This is achieved by a configuration of retinal components that minimizes light scatter while maximizing light sensitivity. The central area, or foveola, consists purely of vertically arranged elongated cones separated from each other by the Originally received: November 29, 1998. Revision accepted: March 19, 1999. Manuscript no. 98761. 1 Department of Ophthalmology and Visual Science, The University of Texas Houston Medical School, Houston, Texas. 2 Department of Ophthalmology, Baylor College of Medicine, Houston, Texas. Presented in part as a poster at the American Academy of Ophthalmology annual meeting, New Orleans, Louisiana, November 1998. Supported in part by unrestricted grants from the Hennann Eye Fund and the J. M. West Texas Corp., Houston, Texas (HMH), Research to Prevent Blindness, Inc., New York, New York (HMH and FLK), NIH grant EYO2520 (FLK), and Vision Core NIH grant EY10608 (HMH). Reprint requests to Helen A. Mintz–Hittner, MD, Department of Ophthalmology and Visual Science, The University of Texas Houston Medical School, 6410 Fannin, Suite 920, Houston, TX 77030-5204. E-mail: [email protected].
processes of the radial fibers of Mu¨ller cells.1 The foveal avascular zone (FAZ), as measured by fluorescein angiography, is the central capillary-free zone of the fovea and has a diameter of 500 to 600 m.2 There have been reports in the literature citing cases in which the FAZ was absent, or traversed by small vessels, both in patients with no known pathology and normal visual acuity3,4 and in patients with macular pathology with normal or decreased visual acuity.5– 8 When evaluating the correlation between retinopathy of prematurity (ROP) severity and immaturity parameters in a previously published study, the observation was made that some of the smallest preterm infants, even if they had developed minimal ROP, had essentially no FAZ on fluorescein angiograms done between 4 and 17 years of age.9 This unusual finding contrasted with histologic studies documenting foveal development that showed that the fovea reaches maturity between 15 and 45 months of age.10 The current study documents the relationship between gestational age and birth weight and the area and diameter of the FAZ on fluorescein angiography in former preterm and term infants.
1409
Ophthalmology Volume 106, Number 7, July 1999
Figure 1. A, fluorescein angiogram of a child who weighed 3000 g and was 40 weeks gestational age showing a normal foveal avascular zone. B, the same fluorescein angiogram with the foveal avascular zone traced out.
Patients and Methods Fundus fluorescein angiograms were performed on both eyes of 150 children aged 1 to 17 years who were 23 to 42 weeks gestational age. Approval was obtained from the Committee for the Protection of Human Subjects to perform fluorescein angiograms on these subjects while they were under general anesthesia for elective surgical procedures, (e.g., strabismus, hernia or circumcision). In order to avoid errors due to fluctuations of vascular endothelial growth factor, patients who had required treatment for active ROP and even those with active ROP ⬎stage 3 mild in either eye were excluded. To avoid errors due to three-dimensional changes in macular positioning, eyes with macular ectopia and macular folds were excluded. To avoid errors due to differences in magnification, eyes with refractive error (expressed as the spherical equivalent) ⬎⫾ five diopters were excluded. Patients who had been large or small for gestational age or who had genetic disorders were excluded. Thus, 101 children were excluded by these criteria. During anesthesia, the patients were placed in the lateral position and a Kowa RC-2 (Kowa Co., Tokyo, Japan) fundus camera (45° angle) was used to photograph the posterior pole through the dilated pupil. Two and one-half ml of 10% sodium fluorescein sterile solution was given by rapid intravenous injection, and a fundus fluorescein angiogram was done on one eye. After 10 minutes, another 2.5 ml of fluorescein was given
by rapid intravenous injection, and a fundus fluorescein angiogram was done on the other eye. The masked unmounted 35-mm fluorescein angiogram negatives were viewed on a light box with an 8⫻ magnifying lens. The frame with the highest quality capillary phase angiogram was selected for one eye of each patient. The negatives were digitized as gray scale images, with the scanner resolution set at 750 pixels per inch using a Polaroid SprintScan 35 Slide Scanner (Polaroid Corp., Cambridge, MA) and saved as bitmap files. Image analysis was done with Jandel Sigma Scan Image Measurement Software version 3.0 (Jandel Scientific, San Rafael, CA). The measurement software was calibrated using a standard image selected randomly from the frames included in the study. The vertical diameter of the optic disc on the standard image was measured in pixels, and the size of 1.64 mm was applied to this distance based on measurements of vertical optic disc diameter in children aged 2 to 10 years by Rimmer et al (Fig 1).11 Each image was viewed on a computer monitor at 300% magnification, and the FAZ was traced. In cases where the central macula appeared to be occupied by a meshwork of capillaries, the location of the fovea was determined by matching the foveal reflex on the color photograph. The area (m2) and greatest diameter (m) were measured for each image. Multiple linear regression analysis was used to evaluate the relationship between the area and greatest diameter of the FAZ, and gestational age, birth weight, gender, race, multiple birth, age at angiography, and so forth.
Results The study group of 49 patients consisted of 12 blacks, 12 Hispanics, and 25 whites. There were 39 children ages 1.0 to 15.5 years who were former preterm infants (ⱕ37 weeks gestational age) and 10 children ages 1.5 to 17.1 years who were former term infants (⬎37 and ⬍42 weeks gestational age). Table 1 lists the characteristics of the former preterm and term infants. Examples of patients with normal, small, and no FAZ are shown in Figures 1, 2, and 3, respectively. The mean FAZ area was 44 m2 (range, 9 to 239 m2; standard deviation [SD] ⫾ 52 m2) in children who were former preterm infants (n ⫽ 39) and 242 m2 (range, 159 to
Table 1. Characteristics of 49 Infants Included in this Study Former Preterm Infants
Former Term Infants
Characteristic
Range
Mean
Range
Mean
FAZ area (m ) FAZ greatest diameter (m) Gestational age (wks) Birth weight (g) ROP stage (ICROP*) Cycloplegic refraction (D†) Age at angiography (yrs) Visual acuity (logMAR‡)
9–239 128–664 23–36 520–2700 0–3* ⫺4.50–⫹3.50 1.0–15.5 ⫺0.30–0.00
44 ⫾ 52 280 ⫾ 122 27.8 ⫾ 3.1 993.5 ⫾ 426.5 2.26 ⫾ 0.64 ⫺0.07 ⫾ 1.76 5.5 ⫾ 3.25 ⫺0.17 ⫾ 0.10
159–288 559–713 38–42 2665–4080 NA 0.00–⫹4.25 1.5–17.1 ⫺0.18–0.00
242 ⫾ 38 652 ⫾ 47 40.1 ⫾ 1.0 3238.0 ⫾ 410.1 NA ⫹1.49 ⫾ 1.49 7.0 ⫾ 5.75 ⫺0.09 ⫾ 0.09
2
NA ⫽ not applicable; FAZ ⫽ foveal avascular zone; ROP ⫽ retinopathy of prematurity. * Stages 0, 1, and 2 are ICROP stages 0, 1, and 2; stage 3 is ICROP stage 3 mild. † Spherical equivalents were measured during the office visit just prior to the fluorescein angiogram. ‡ Visual acuities were measured by Allen figures or Snellen test types during the office visit just prior to the fluorescein angiogram (n ⫽ 41) or by Allen figures as soon as the infant became verbal (n ⫽ 8).
1410
Mintz–Hittner et al 䡠 Small FAZ May Be an Historic Mark of Prematurity diameter are plotted against gestational age and birth weight in Figure 4.
Discussion
Figure 2. Fluorescein angiograms of children showing a small foveal avascular zone. A, child who weighed 1000 g and was 29 weeks gestational age. B, child who weighed 1645 g and was 32 weeks gestational age.
288 m2; SD ⫾ 38 m2) in children who were former term infants (n ⫽ 10). The mean FAZ greatest diameter was 280 m (range, 128 to 664 m; SD ⫾ 122 m) in children who were former preterm infants (n ⫽ 39) and 652 m (range, 559 to 713 m; SD ⫾47 m) in children who were former term infants (n ⫽ 10). Multiple linear regression analysis identified only gestational age and birth weight as significant correlations (P ⬍0.0001). Foveal avascular zone area and greatest
This study documented that, in children, FAZ area and diameter correlated with gestational age and birth weight but not with any of the other parameters studied. The measurements obtained in this study correlated well with published reports of normal FAZ size, i.e., approximately 500 to 600 m diameter. The formation of the FAZ during vasculogenesis has been explained by two opposing hypotheses. Henkind et al12 maintained that the central macula (in kittens and the Rhesus monkey) is fully vascularized in fetal life and that subsequent vascular remodeling results in the formation of the FAZ. In contrast, Engerman13 found that the primitive macular capillaries (in the Macaca mulatta monkey) ceased to proliferate once reaching the foveal center. This study supports Henkind’s hypothesis. The process of vascular remodeling of the fovea by apoptosis14 can be compared to the changes that occur
Figure 3. Fluorescein angiograms of children showing absence of a foveal avascular zone. A, child who weighed 965 g and was 28 weeks gestational age. B, child who weighed 600 g and was 24 weeks gestational age. C, child who weighed 750 g and was 25 weeks gestational age. D, child who weighed 625 g and was 26 weeks gestational age.
1411
Ophthalmology Volume 106, Number 7, July 1999
Figure 4. The correlations between area and greatest diameter of the foveal avascular zone and gestational age and birth weight. A, FAZ area (m2) plotted against gestational age (weeks) (R/F/P ⫽ 0.88/166.70/⬍0.0001). B, FAZ greatest diameter (m) plotted against gestational age (weeks) (R/F/P ⫽ 0.87/ 151.10/⬍0.0001). C, FAZ area (m2) plotted against birth weight (g) (R/F/P ⫽ 0.88/167.06/⬍0.0001). D, FAZ greatest diameter (m) plotted against birth weight (g) (R/F/P ⫽ 0.87/148.74/ ⬍0.0001).
sequentially in other vascular structures during ocular development, such as the tunica vasculosa lentis (TVL).15 The TVL covers the entire lens until 28 weeks, regresses gradually between 29 and 34 weeks, and is absent after 35 weeks.16 Recent reports have implicated macrophages in the induction of apoptosis of vascular endothelial cells during the regression of the pupillary membrane (TVL) in the rat.15 This appears to be regulated by vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (FGF) levels.17 Remnants of the TVL may occur in various degrees, ranging from an isolated insignificant strand, without visual consequence,16 to an hereditary dense pupillary membrane, resulting in marked obstruction of vision.18 It is probable that the same mechanism of regression by apoptosis takes place in the fovea to form the FAZ, except that the exact time sequence seems to be less precise and the role of VEGF or FGF has not been established. Incomplete regression seems to result in the presence of vessels traversing the FAZ in otherwise normal eyes3,4 or in eyes with pathology.5–9 Thus, apoptosis is a basic biologic phenomenon with wide-ranging implications in tissue kinetics14 and is likely the mechanism of regression for both embryonic ocular vascular systems (the TVL of the hyaloid and the FAZ of the inner retinal vascular systems).15 Even in infants who are born very prematurely and develop severe ROP, the TVL usually regresses, although this may be delayed and remnants may remain. In contrast, in infants who are born before 30 weeks gestational age, even if they do not develop any significant ROP, a normal FAZ usually fails to form. Interestingly, vascular remodeling in the fovea is not required for the development of good visual acuity. Thus, although there is no effect on
1412
final visual acuity, a small or absent FAZ may be an historic mark of prematurity.
References 1. Yamada E. Some structural features of the fovea centralis in the human retina. Arch Ophthalmol 1969;82:151–9. 2. Early Treatment for Diabetic Retinopathy Study Research Group. Classification of diabetic retinopathy from fluorescein angiograms. ETDRS report number 11. Ophthalmology 1991; 98(5 Suppl):807–22. 3. Bird AC, Weale RA. On the retinal vasculature of the human fovea. Exp Eye Res 1974;19:409 –17. 4. Yeung J, Crock G, Cairns J, et al. Macular-foveal capillaries in human retina. Aust J Ophthalmol 1973;1:17–23. 5. Summers CG, Knobloch WH, Witkop CJ Jr, King RA. Hermansky-Pudlak syndrome. Ophthalmic findings. Ophthalmology 1988;95:545–54. 6. Mintz-Hittner HA, Ferrell RE, Lyons LA, Kretzer FL. Criteria to detect minimal expressivity within families with autosomal dominant aniridia. Am J Ophthalmol 1992;114: 700 –7. 7. Goldberg MF, Custis PH. Retinal and other manifestations of incontinentia pigmenti (Bloch-Sulzberger syndrome). Ophthalmology 1993;100:1645–54. 8. Ibayashi H, Nishimura M, Yamana T. Avascular zone in the macula in cicatricial retinopathy of prematurity. Am J Ophthalmol 1985;99:235–9. 9. Mintz-Hittner HA, Prager TC, Kretzer FL. Visual acuity correlates with severity of retinopathy of prematurity in untreated infants weighing 750 g or less at birth. Arch Ophthalmol 1992;110:1087–91. 10. Hendrickson AE, Yuodelis C. The morphological development of the human fovea. Ophthalmology 1984;91:603– 12.
Mintz–Hittner et al 䡠 Small FAZ May Be an Historic Mark of Prematurity 11. Rimmer S, Keating C, Chou T, et al. Growth of the human optic disk and nerve during gestation, childhood, and early adulthood. Am J Ophthalmol 1993;116:748 – 53. 12. Henkind P, Bellhorn RW, Murphy ME, Roa N. Development of macular vessels in monkey and cat. Br J Ophthalmol 1975;59:703–9. 13. Engerman RL. Development of the macular circulation. Invest Ophthalmol 1976;15:835– 40. 14. Kerr JFR, Wyllie AR, Currie AR. Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer 1972;26:239 –57.
15. Lang R, Lustig M, Francois F, et al. Apoptosis during macrophage-dependent ocular tissue remodelling. Development 1994;120:3395– 403. 16. Hittner HM, Hirsch NJ, Rudolph AJ. Assessment of gestational age by examination of the anterior vascular capsule of the lens. J Pediatr 1977;91:455– 8. 17. Yanagawa T, Matsuo T, Matsuo N. Aqueous vascular endothelial growth factor and basic fibroblast growth factor decrease during regression of rabbit pupillary membrane. Jpn J Ophthalmol 1998;42:157– 61. 18. Cassady JR, Light A. Familial persistent pupillary membranes. Arch Ophthalmol 1957;58:438 – 48.
1413
processes of the radial fibers of Mu¨ller cells.1 The foveal avascular zone (FAZ), as measured by fluorescein angiography, is the central capillary-free zone of the fovea and has a diameter of 500 to 600 m.2 There have been reports in the literature citing cases in which the FAZ was absent, or traversed by small vessels, both in patients with no known pathology and normal visual acuity3,4 and in patients with macular pathology with normal or decreased visual acuity.5– 8 When evaluating the correlation between retinopathy of prematurity (ROP) severity and immaturity parameters in a previously published study, the observation was made that some of the smallest preterm infants, even if they had developed minimal ROP, had essentially no FAZ on fluorescein angiograms done between 4 and 17 years of age.9 This unusual finding contrasted with histologic studies documenting foveal development that showed that the fovea reaches maturity between 15 and 45 months of age.10 The current study documents the relationship between gestational age and birth weight and the area and diameter of the FAZ on fluorescein angiography in former preterm and term infants.
1409
Ophthalmology Volume 106, Number 7, July 1999
Figure 1. A, fluorescein angiogram of a child who weighed 3000 g and was 40 weeks gestational age showing a normal foveal avascular zone. B, the same fluorescein angiogram with the foveal avascular zone traced out.
Patients and Methods Fundus fluorescein angiograms were performed on both eyes of 150 children aged 1 to 17 years who were 23 to 42 weeks gestational age. Approval was obtained from the Committee for the Protection of Human Subjects to perform fluorescein angiograms on these subjects while they were under general anesthesia for elective surgical procedures, (e.g., strabismus, hernia or circumcision). In order to avoid errors due to fluctuations of vascular endothelial growth factor, patients who had required treatment for active ROP and even those with active ROP ⬎stage 3 mild in either eye were excluded. To avoid errors due to three-dimensional changes in macular positioning, eyes with macular ectopia and macular folds were excluded. To avoid errors due to differences in magnification, eyes with refractive error (expressed as the spherical equivalent) ⬎⫾ five diopters were excluded. Patients who had been large or small for gestational age or who had genetic disorders were excluded. Thus, 101 children were excluded by these criteria. During anesthesia, the patients were placed in the lateral position and a Kowa RC-2 (Kowa Co., Tokyo, Japan) fundus camera (45° angle) was used to photograph the posterior pole through the dilated pupil. Two and one-half ml of 10% sodium fluorescein sterile solution was given by rapid intravenous injection, and a fundus fluorescein angiogram was done on one eye. After 10 minutes, another 2.5 ml of fluorescein was given
by rapid intravenous injection, and a fundus fluorescein angiogram was done on the other eye. The masked unmounted 35-mm fluorescein angiogram negatives were viewed on a light box with an 8⫻ magnifying lens. The frame with the highest quality capillary phase angiogram was selected for one eye of each patient. The negatives were digitized as gray scale images, with the scanner resolution set at 750 pixels per inch using a Polaroid SprintScan 35 Slide Scanner (Polaroid Corp., Cambridge, MA) and saved as bitmap files. Image analysis was done with Jandel Sigma Scan Image Measurement Software version 3.0 (Jandel Scientific, San Rafael, CA). The measurement software was calibrated using a standard image selected randomly from the frames included in the study. The vertical diameter of the optic disc on the standard image was measured in pixels, and the size of 1.64 mm was applied to this distance based on measurements of vertical optic disc diameter in children aged 2 to 10 years by Rimmer et al (Fig 1).11 Each image was viewed on a computer monitor at 300% magnification, and the FAZ was traced. In cases where the central macula appeared to be occupied by a meshwork of capillaries, the location of the fovea was determined by matching the foveal reflex on the color photograph. The area (m2) and greatest diameter (m) were measured for each image. Multiple linear regression analysis was used to evaluate the relationship between the area and greatest diameter of the FAZ, and gestational age, birth weight, gender, race, multiple birth, age at angiography, and so forth.
Results The study group of 49 patients consisted of 12 blacks, 12 Hispanics, and 25 whites. There were 39 children ages 1.0 to 15.5 years who were former preterm infants (ⱕ37 weeks gestational age) and 10 children ages 1.5 to 17.1 years who were former term infants (⬎37 and ⬍42 weeks gestational age). Table 1 lists the characteristics of the former preterm and term infants. Examples of patients with normal, small, and no FAZ are shown in Figures 1, 2, and 3, respectively. The mean FAZ area was 44 m2 (range, 9 to 239 m2; standard deviation [SD] ⫾ 52 m2) in children who were former preterm infants (n ⫽ 39) and 242 m2 (range, 159 to
Table 1. Characteristics of 49 Infants Included in this Study Former Preterm Infants
Former Term Infants
Characteristic
Range
Mean
Range
Mean
FAZ area (m ) FAZ greatest diameter (m) Gestational age (wks) Birth weight (g) ROP stage (ICROP*) Cycloplegic refraction (D†) Age at angiography (yrs) Visual acuity (logMAR‡)
9–239 128–664 23–36 520–2700 0–3* ⫺4.50–⫹3.50 1.0–15.5 ⫺0.30–0.00
44 ⫾ 52 280 ⫾ 122 27.8 ⫾ 3.1 993.5 ⫾ 426.5 2.26 ⫾ 0.64 ⫺0.07 ⫾ 1.76 5.5 ⫾ 3.25 ⫺0.17 ⫾ 0.10
159–288 559–713 38–42 2665–4080 NA 0.00–⫹4.25 1.5–17.1 ⫺0.18–0.00
242 ⫾ 38 652 ⫾ 47 40.1 ⫾ 1.0 3238.0 ⫾ 410.1 NA ⫹1.49 ⫾ 1.49 7.0 ⫾ 5.75 ⫺0.09 ⫾ 0.09
2
NA ⫽ not applicable; FAZ ⫽ foveal avascular zone; ROP ⫽ retinopathy of prematurity. * Stages 0, 1, and 2 are ICROP stages 0, 1, and 2; stage 3 is ICROP stage 3 mild. † Spherical equivalents were measured during the office visit just prior to the fluorescein angiogram. ‡ Visual acuities were measured by Allen figures or Snellen test types during the office visit just prior to the fluorescein angiogram (n ⫽ 41) or by Allen figures as soon as the infant became verbal (n ⫽ 8).
1410
Mintz–Hittner et al 䡠 Small FAZ May Be an Historic Mark of Prematurity diameter are plotted against gestational age and birth weight in Figure 4.
Discussion
Figure 2. Fluorescein angiograms of children showing a small foveal avascular zone. A, child who weighed 1000 g and was 29 weeks gestational age. B, child who weighed 1645 g and was 32 weeks gestational age.
288 m2; SD ⫾ 38 m2) in children who were former term infants (n ⫽ 10). The mean FAZ greatest diameter was 280 m (range, 128 to 664 m; SD ⫾ 122 m) in children who were former preterm infants (n ⫽ 39) and 652 m (range, 559 to 713 m; SD ⫾47 m) in children who were former term infants (n ⫽ 10). Multiple linear regression analysis identified only gestational age and birth weight as significant correlations (P ⬍0.0001). Foveal avascular zone area and greatest
This study documented that, in children, FAZ area and diameter correlated with gestational age and birth weight but not with any of the other parameters studied. The measurements obtained in this study correlated well with published reports of normal FAZ size, i.e., approximately 500 to 600 m diameter. The formation of the FAZ during vasculogenesis has been explained by two opposing hypotheses. Henkind et al12 maintained that the central macula (in kittens and the Rhesus monkey) is fully vascularized in fetal life and that subsequent vascular remodeling results in the formation of the FAZ. In contrast, Engerman13 found that the primitive macular capillaries (in the Macaca mulatta monkey) ceased to proliferate once reaching the foveal center. This study supports Henkind’s hypothesis. The process of vascular remodeling of the fovea by apoptosis14 can be compared to the changes that occur
Figure 3. Fluorescein angiograms of children showing absence of a foveal avascular zone. A, child who weighed 965 g and was 28 weeks gestational age. B, child who weighed 600 g and was 24 weeks gestational age. C, child who weighed 750 g and was 25 weeks gestational age. D, child who weighed 625 g and was 26 weeks gestational age.
1411
Ophthalmology Volume 106, Number 7, July 1999
Figure 4. The correlations between area and greatest diameter of the foveal avascular zone and gestational age and birth weight. A, FAZ area (m2) plotted against gestational age (weeks) (R/F/P ⫽ 0.88/166.70/⬍0.0001). B, FAZ greatest diameter (m) plotted against gestational age (weeks) (R/F/P ⫽ 0.87/ 151.10/⬍0.0001). C, FAZ area (m2) plotted against birth weight (g) (R/F/P ⫽ 0.88/167.06/⬍0.0001). D, FAZ greatest diameter (m) plotted against birth weight (g) (R/F/P ⫽ 0.87/148.74/ ⬍0.0001).
sequentially in other vascular structures during ocular development, such as the tunica vasculosa lentis (TVL).15 The TVL covers the entire lens until 28 weeks, regresses gradually between 29 and 34 weeks, and is absent after 35 weeks.16 Recent reports have implicated macrophages in the induction of apoptosis of vascular endothelial cells during the regression of the pupillary membrane (TVL) in the rat.15 This appears to be regulated by vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (FGF) levels.17 Remnants of the TVL may occur in various degrees, ranging from an isolated insignificant strand, without visual consequence,16 to an hereditary dense pupillary membrane, resulting in marked obstruction of vision.18 It is probable that the same mechanism of regression by apoptosis takes place in the fovea to form the FAZ, except that the exact time sequence seems to be less precise and the role of VEGF or FGF has not been established. Incomplete regression seems to result in the presence of vessels traversing the FAZ in otherwise normal eyes3,4 or in eyes with pathology.5–9 Thus, apoptosis is a basic biologic phenomenon with wide-ranging implications in tissue kinetics14 and is likely the mechanism of regression for both embryonic ocular vascular systems (the TVL of the hyaloid and the FAZ of the inner retinal vascular systems).15 Even in infants who are born very prematurely and develop severe ROP, the TVL usually regresses, although this may be delayed and remnants may remain. In contrast, in infants who are born before 30 weeks gestational age, even if they do not develop any significant ROP, a normal FAZ usually fails to form. Interestingly, vascular remodeling in the fovea is not required for the development of good visual acuity. Thus, although there is no effect on
1412
final visual acuity, a small or absent FAZ may be an historic mark of prematurity.
References 1. Yamada E. Some structural features of the fovea centralis in the human retina. Arch Ophthalmol 1969;82:151–9. 2. Early Treatment for Diabetic Retinopathy Study Research Group. Classification of diabetic retinopathy from fluorescein angiograms. ETDRS report number 11. Ophthalmology 1991; 98(5 Suppl):807–22. 3. Bird AC, Weale RA. On the retinal vasculature of the human fovea. Exp Eye Res 1974;19:409 –17. 4. Yeung J, Crock G, Cairns J, et al. Macular-foveal capillaries in human retina. Aust J Ophthalmol 1973;1:17–23. 5. Summers CG, Knobloch WH, Witkop CJ Jr, King RA. Hermansky-Pudlak syndrome. Ophthalmic findings. Ophthalmology 1988;95:545–54. 6. Mintz-Hittner HA, Ferrell RE, Lyons LA, Kretzer FL. Criteria to detect minimal expressivity within families with autosomal dominant aniridia. Am J Ophthalmol 1992;114: 700 –7. 7. Goldberg MF, Custis PH. Retinal and other manifestations of incontinentia pigmenti (Bloch-Sulzberger syndrome). Ophthalmology 1993;100:1645–54. 8. Ibayashi H, Nishimura M, Yamana T. Avascular zone in the macula in cicatricial retinopathy of prematurity. Am J Ophthalmol 1985;99:235–9. 9. Mintz-Hittner HA, Prager TC, Kretzer FL. Visual acuity correlates with severity of retinopathy of prematurity in untreated infants weighing 750 g or less at birth. Arch Ophthalmol 1992;110:1087–91. 10. Hendrickson AE, Yuodelis C. The morphological development of the human fovea. Ophthalmology 1984;91:603– 12.
Mintz–Hittner et al 䡠 Small FAZ May Be an Historic Mark of Prematurity 11. Rimmer S, Keating C, Chou T, et al. Growth of the human optic disk and nerve during gestation, childhood, and early adulthood. Am J Ophthalmol 1993;116:748 – 53. 12. Henkind P, Bellhorn RW, Murphy ME, Roa N. Development of macular vessels in monkey and cat. Br J Ophthalmol 1975;59:703–9. 13. Engerman RL. Development of the macular circulation. Invest Ophthalmol 1976;15:835– 40. 14. Kerr JFR, Wyllie AR, Currie AR. Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer 1972;26:239 –57.
15. Lang R, Lustig M, Francois F, et al. Apoptosis during macrophage-dependent ocular tissue remodelling. Development 1994;120:3395– 403. 16. Hittner HM, Hirsch NJ, Rudolph AJ. Assessment of gestational age by examination of the anterior vascular capsule of the lens. J Pediatr 1977;91:455– 8. 17. Yanagawa T, Matsuo T, Matsuo N. Aqueous vascular endothelial growth factor and basic fibroblast growth factor decrease during regression of rabbit pupillary membrane. Jpn J Ophthalmol 1998;42:157– 61. 18. Cassady JR, Light A. Familial persistent pupillary membranes. Arch Ophthalmol 1957;58:438 – 48.
1413