Pediatric Annals

Retinopathy of Prematurity: Progress Report

Gordon B Avery, MD, PhD; Penny Glass, PhD

Abstract

What ever happened to retrolental fibroplasia? This chapter of medical history is both embarrassing and educational. The development of efficient incubators ushered in the modem era of nursery care, with increased survival of small premature infants. Respiratory problems were common in these small babies, and oxygen was shown to benefit many. In addition, apneic spells were reduced when premature infants were given supplementary oxygen. By the mid-1940s, it was considered good practice to supply oxygen routinely to all prematures, the more the better. The result is well known: retrolental fibroplasia (RLF) became the leading cause of blindness and thousands of blind babies were discharged from premature nurseries in the 1940s and 1950s.

When it was finally appreciated that oxygen could be toxic to the immature retina, oxygen use was drastically curtailed. Hypoxie problems followed hyperoxic ones, and it has been estimated that 16 babies died or suffered hypoxic brain damage for each infant saved from RLF.1 When the wild swinging of the pendulum ceased, oxygen was understood to be both lifesaving and toxic. Since that time, oxygen has been carefully adjusted to attempt to provide arterial oxygen tensions in the physiologic range. However, it has proved impossible to keep arterial oxygen at precisely designated levels in labile, sick prematures. Furthermore, any selected oxygen value is a compromise, with lower oxygen tensions which are normal for the fetal retina being untenable for the postnatal premature baby.

With controlled oxygen administration, retinopathy of prematurity was relatively rare until the last decade or so, when significant numbers of prematures less than 1,000 g began to survive. The new name, retinopathy of prematurity (ROP), recognizes that immaturity at birth is the single largest risk factor for this disease. Furthermore, of infants with ROP, only a minority develop the scarring disease with retinal detachment originally called retrolental fibroplasia. A number of additional risk factors combine with retinal immaturity to contribute to the current incidence of the disease. It is generally considered that ROP, after all these years of study, is a disease of increasing prevalence, varying severity, and multiple causation.

DEFINITION AND PROGNOSIS

Alteration of normal retinal vascular development is the hallmark of retinopathy of prematurity. Low birth weight infants and infants exposed to supplementary oxygen should have their eyes examined by a pediatrie ophthalmologist by 6 to 8 weeks of age. Examination of the preterm retina by indirect ophthalmoscope will reveal one of the following conditions:

1. Mature vascularization, complete to the ora serrata, and no ROP.

2. Immature vascularization, with the appearance of a gradual transition between the vascular and avascular zone, still at risk for developing ROP until vascularization is complete.

3. ROP, either active or regressing. Management by a pediatrie ophthalmologist is advised. The prognosis depends on the severity (stage 1-4), the location (zone 1-3), and the extent of the disease (1-12 clock hours).2 Zone 1 (posterior) has a poorer prognosis. The appearance of "plus disease , " or vascular engorgement and tortuosity, is an additional marker of severity.

In stage J, a demarcation between the vascular and avascular zone appears, which represents early "shunt" formation. By stage 2, the shunt has elevated above the plane of the retina. Most ROP arrests at stage 1 or 2 and regresses; progression to higher grades is unpredictable. Infants with stage 1 or 2 disease may have a retinal scar at the location of the former shunt or the disease may not even be detectable on later examination. However, even regressed ROP is associated with an increased risk for refractive errors, amblyopia, and strabismus.

Stage 3 is defined by anomalous neovascularization into the vitreous…

What ever happened to retrolental fibroplasia? This chapter of medical history is both embarrassing and educational. The development of efficient incubators ushered in the modem era of nursery care, with increased survival of small premature infants. Respiratory problems were common in these small babies, and oxygen was shown to benefit many. In addition, apneic spells were reduced when premature infants were given supplementary oxygen. By the mid-1940s, it was considered good practice to supply oxygen routinely to all prematures, the more the better. The result is well known: retrolental fibroplasia (RLF) became the leading cause of blindness and thousands of blind babies were discharged from premature nurseries in the 1940s and 1950s.

When it was finally appreciated that oxygen could be toxic to the immature retina, oxygen use was drastically curtailed. Hypoxie problems followed hyperoxic ones, and it has been estimated that 16 babies died or suffered hypoxic brain damage for each infant saved from RLF.1 When the wild swinging of the pendulum ceased, oxygen was understood to be both lifesaving and toxic. Since that time, oxygen has been carefully adjusted to attempt to provide arterial oxygen tensions in the physiologic range. However, it has proved impossible to keep arterial oxygen at precisely designated levels in labile, sick prematures. Furthermore, any selected oxygen value is a compromise, with lower oxygen tensions which are normal for the fetal retina being untenable for the postnatal premature baby.

With controlled oxygen administration, retinopathy of prematurity was relatively rare until the last decade or so, when significant numbers of prematures less than 1,000 g began to survive. The new name, retinopathy of prematurity (ROP), recognizes that immaturity at birth is the single largest risk factor for this disease. Furthermore, of infants with ROP, only a minority develop the scarring disease with retinal detachment originally called retrolental fibroplasia. A number of additional risk factors combine with retinal immaturity to contribute to the current incidence of the disease. It is generally considered that ROP, after all these years of study, is a disease of increasing prevalence, varying severity, and multiple causation.

DEFINITION AND PROGNOSIS

Alteration of normal retinal vascular development is the hallmark of retinopathy of prematurity. Low birth weight infants and infants exposed to supplementary oxygen should have their eyes examined by a pediatrie ophthalmologist by 6 to 8 weeks of age. Examination of the preterm retina by indirect ophthalmoscope will reveal one of the following conditions:

1. Mature vascularization, complete to the ora serrata, and no ROP.

2. Immature vascularization, with the appearance of a gradual transition between the vascular and avascular zone, still at risk for developing ROP until vascularization is complete.

3. ROP, either active or regressing. Management by a pediatrie ophthalmologist is advised. The prognosis depends on the severity (stage 1-4), the location (zone 1-3), and the extent of the disease (1-12 clock hours).2 Zone 1 (posterior) has a poorer prognosis. The appearance of "plus disease , " or vascular engorgement and tortuosity, is an additional marker of severity.

In stage J, a demarcation between the vascular and avascular zone appears, which represents early "shunt" formation. By stage 2, the shunt has elevated above the plane of the retina. Most ROP arrests at stage 1 or 2 and regresses; progression to higher grades is unpredictable. Infants with stage 1 or 2 disease may have a retinal scar at the location of the former shunt or the disease may not even be detectable on later examination. However, even regressed ROP is associated with an increased risk for refractive errors, amblyopia, and strabismus.

Stage 3 is defined by anomalous neovascularization into the vitreous which may produce scarring and traction on the retina or dragging of the disk. These infants typically have a high refractive error and require corrective lenses. Strabismus is common. Infants who had retinal scarring are subject to late retinal detachment. In selected cases, cryotherapy is being applied to prevent this complication. A recently completed multicenter clinical trial demonstrated a reduction in retinal detachment for eyes treated with cryotherapy versus untreated eyes that had reached a defined threshold. * The long term outcome following treatment is not yet known. The study group recommended treatment of one eye if both eyes are affected.

The term "retrolental fibroplasia" is now usually reserved for stage 4 ROP, retinal detachment. The detachment may be partial or complete and the infant may be blind or have minimal functional vision. Later complications may include acute angle-closure glaucoma. Surgery may be necessary to save the globe. Attempts to reattach the retina surgically have met with limited success. The patient may still be functionally blind. Parents need support from their pediarrician and community vision services to help them cope with rheir child's handicap.

Table

TABLERisk Factors

TABLE

Risk Factors

INCIDENCE

Retinopathy of prematurity is almost exclusively a disease of the very low birth weight infant. The majority of carefully observed prematures below 1,000 g birth weight have at least ROP ot lower severity. The Figure shows the overall incidence of ROP by birth weight. Phelps has estimated that 367 prematures annually would survive with scarring ROP from the birth weight group 1,000 to 1,500 g, and 749 with scarring retinopathy would survive from the group with birth weight of less than 1,000 g.4 Based on these data, approximately 600 infants each year are blind as a result of ROP.

THEORIES OF CAUSATION

Although a great deal ot effort has been invested in clinical and animal investigations of ROP, we are left with an etiological labyrinth in which no single factor stands alone. The Table lists factors which have been epidemiologically linked to ROP in one or more reports. 5 One is forced to conclude that the emhryonic retina of the small premature, developing outside the uterus, is vulnerable to many sources of disturbance which can disrupt orderly differentiation and vascularization. ROP is thus the final common pathway hy which this disrupted development is expressed. A formula for ROP could he:

Figure. The first bar in each group shows the incidence of ROP for various birih weight ranges. The second bar shows the proportion of total births in each birth weight range. It is clear that most births occur at >2,000 grams while most ROP occurs at < 1,000 grams. Data from Glass P Avery GB. Subramanian KNS, et al; New Eng! J Med 1985; 313.401-404.

Figure. The first bar in each group shows the incidence of ROP for various birih weight ranges. The second bar shows the proportion of total births in each birth weight range. It is clear that most births occur at >2,000 grams while most ROP occurs at < 1,000 grams. Data from Glass P Avery GB. Subramanian KNS, et al; New Eng! J Med 1985; 313.401-404.

Immaturity (always) + Oxygen (often) + Other Factors (variably) = ROP

Besides the well known impact of hyperoxia, it appears that hypoxia, variations in PaCO2, pH, retinal oxygen consumption, light exposure, and other factors that affect retinal perfusion all may play a role. Kitten and other animal models support the impact of many of these factors. Limitations of space preclude a detailed review here, hut we have recently summarized the available literature. 5 Perhaps a unifying hypothesis is that the smallest, sickest babies have the most complications which disrupt retinal perfusion and metabolism. They also have the most ROP.

SUMMARY

For the time being, risk of ROP seems inexorably linked to survival of extremely low birth weight prematures, whose emhryonic retinas develop in an abnormal and fluctuating environment. Incidence and severity may be reduced by stabilizing ventilation, oxygenation, and perfusion, moderating light exposure, and providing normal levels of vitamin E. Progression of stage 3 retinopathy may sometimes be arrested by cryotherapy. In the future, antioxidants or other pharmacologie agents may be developed to provide a greater margin of safety. In the meantime, the eyes of prematures must be examined and those with ROP will need specialized ophthalmologic care.

REFERENCES

1. Cross WE: Cost of preventing retrolental fibroplasia. Lancet 1973; 2:954.

2. Special Committee Report: An international classification of retinopathy of prematurity. Pediatrics 1984; 74:127-133.

3. Cryotheraphy for Retinopathy of Prematurity Group: Multicenter trial of cryotherapy for retinopathy of prematurity: Preliminary results. Pediatrics 1988; 81:697-706.

4. Phelps D: Vitamin E in retinopathy of prematurity, in Silverman WA, Flynn JT: Retinopathy of Prematurity. Bodston, Blackwell Scientific Publications, 1985, p 199.

5. Avery GB, Glass P: Retinopathy of prematurity: What causes it? Clin Perinatol, Dec, 1988.

TABLE

Risk Factors

10.3928/0090-4481-19880801-10

Sign up to receive

Journal E-contents