- Top of page
- Materials and Methods
Background: To report the structural and refractive outcome after laser photocoagulation for retinopathy of prematurity (ROP).
Methods: Nineteen consecutive patients who had undergone photocoagulation for ROP between 1997 and 2002 at our clinic were examined for this non-comparative, consecutive, interventional, retrospective case series. A total of 37 eyes received either transscleral or transpupillary laser treatment. Data consisted of grade of ROP pre- and postoperatively, birth weight, perioperative and postoperative complications and refraction. Based on indirect ophthalmoscopy, independent observers graded the extent of ROP and determined the postoperative refraction by retinoscopy.
Results: A total of 97% of all eyes responded to laser treatment with regression of ROP. Only one eye out of 37 progressed to stage IV B despite photocoagulation and therefore an encircling procedure was performed. After further progression a vitrectomy was carried out. Perioperative complications included haemorrhages in 22% that resorbed spontaneously and cataract formation in one eye (3%). Postoperative refractive errors at mean ages of 23 ± 12 months and 45 ± 14 months were evaluated in 15/19 patients (79%). The spherical equivalents ranged between −8 D and +6 D at the first examination and between −12 D and +7 D at the second examination. In all only 14% of the refracted eyes were myopic.
Conclusions: Photocoagulation for ROP in our patients resulted in regression of threshold ROP. In addition, the analyses of the refractive outcomes demonstrated a predominance of hypermetropia in our patients.
- Top of page
- Materials and Methods
Retinopathy of prematurity (ROP) represents a disease of the retinal vasculature in preterm infants resulting in retinal vascular proliferations (Hunter & Mukai 1992). Terry (1942) named this disorder ‘retrolental fibroplasia’ because of the presence of membranes in the posterior compartment in the late stages of the disease. The pathomechanism of ROP has been found to correlate with the amount of oxygen supplementation given to premature infants and the incidence of retinal changes decreases with reduced oxygen supplementation (Lanman et al. 1954; Kinsey et al. 1977).
Retinopathy of prematurity is now understood as a two- phase process. During the acute first phase, the normal vasculogenesis of the retina is disturbed by the relative hyperoxia of the extra-uterine environment. This causes vaso-obliteration and non-vascularization of some areas of the anterior retina. The subsequent hypoxia induces an up-regulation of vascular endothelial growth factor and other angiogenic factors that play a crucial role in the second chronic phase, characterized by vaso-proliferation (Wheatley et al. 2002; Hutcheson & Kelly 2003).
Laser photocoagulation can be applied via either the transpupillary or the transscleral route. Both techniques have been shown to be equally effective (Seiberth et al. 1997). Transscleral administration, on one hand, offers the advantage of decreasing the risk of cataract formation by circumventing the lens. On the other hand, it may be more traumatic, especially when the conjunctiva has to be incised to reach the areas that need to be treated.
The aims of the present study were firstly to investigate the efficacy of laser treatment in ROP and the related incidence of intra- and postoperative complications and, secondly, to analyse the refractive outcomes of our patients following laser photocoagulation.
Materials and Methods
- Top of page
- Materials and Methods
We reviewed the clinical records of 19 consecutive patients who had undergone laser photocoagulation for threshold ROP at the University Clinic of Ophthalmology in Innsbruck between 1997 and 2002. Threshold ROP was determined as stage III ROP in five or more contiguous or eight cumulative 30-degree sectors (8.4 clock hours) in zones I or II in the presence of ‘plus’ disease (Multicentre Trial of Cryotherapy for Retinopathy of Prematurity 1988). Approval by the IRB was not required for this retrospective case series. A total of 37 eyes were included in the study. Except for one patient who had asymmetrical disease, all patients had received laser therapy to both eyes because of bilateral threshold ROP.
Laser photocoagulation was performed with the patient under general anaesthesia. At the start patients with adequate dilation were treated with transpupillary laser photocoagulation and additional scleral indentation while those patients with inadequate dilation were treated with transscleral photocoagulation. In 23 eyes (62%) transscleral diode laser treatment was applied from the beginning. In 14 eyes we started with indirect laser therapy, but in eight of these 14 eyes we converted to transscleral application because the transscleral technique allows coagulation of the periphery even in the presence of a poorly dilated pupil while sparing the lens. In sum, transscleral laser was applied in 84% of the eyes and transpupillary treatment in 16% of the eyes as our surgeons came to prefer the transscleral technique as the study proceeded. For both laser applications a diode laser with an 810-nm wavelength (IRIS Medical Instruments Inc., Mountain View, California, USA) was used. A conjunctival incision was performed in five eyes.
The average laser power used with the transscleral technique was 453 mW (range 200–800 mW) per eye and the mean number of burns was 175 (range 94–325). With the transpupillary mode the mean number of burns was 1019 per eye with duration of 100 ms. Topical antibiotics (0.3% gentamicin sulfate) and NSAIDs (non-steroidal anti-inflammatory drugs) were applied after surgery.
During chart review, the following preoperative data were obtained for each patient: age, gender, birth weight and the grade of ROP. Postoperative data included complications associated with the laser treatment, grade of ROP, refraction (when possible) and evaluation whether further surgery was necessary due to failure of laser photocoagulation. The anatomic outcome was evaluated by indirect ophthalmoscopy. Unfavourable outcomes were assessed in accordance with the Multicentre Trial of Cryotherapy for ROP (1988). The mean follow-up time was 45 ± 14 months. Only three ophthalmologists, two surgeons and the masked observer who examined the patients in the follow-up period were in charge of the study patients. This enabled us to minimize possible study bias.
- Top of page
- Materials and Methods
In our study the mean birth weight was 816 ± 283 g (range 480−1455 g) and the mean gestational age was 26 ± 2 weeks (range 23−31 weeks. The female : male ratio was 10 female versus nine male infants (Table 1).
Table 1. Preoperative data for all patients.
|Patient||Gestational||Gender||Birth||Stage of ROP|
| ||age (weeks)|| ||weight (grams)||Right eye ||Left eye|
|3||24||F||680||III +||III +|
|4||27||M||1100||III +||III +|
|8||28||M||1230||III +||III +|
|9||24||F||560||III +||III ± IV|
|10||29||F||840||III +||III +|
|11||25||M||510||III +||III +|
|12||24||M||690||III +||III +|
|13||29||F||1390||III +||III +|
|14||25||F||660||III +||III +|
|15||27||F||960||III +||III +|
|16||24||F||670||III +||III +|
|17||31||F||1455||III +||III +|
|18||25||F||730||III +||III +|
|19||25||M||480||III +||III +|
Regression of the disease after laser therapy was observed in 36/37 eyes (97%). Only 6/37 eyes required repetitive laser therapy, while a single treatment proved to be sufficient in 31/37 eyes.
In 35/37 eyes no adverse effects such as burning anterior segment tissues or bleeding in the anterior chamber occurred. Transscleral application induced minimal conjunctival swelling in some of the eyes. No severe posterior segment side-effects were observed except for slight haemorrhages in 22% of the eyes, which were absorbed spontaneously during follow-up.
In three eyes these haemorrhages were located at the ridge, while in five eyes the haemorrhages were located centrally from the ridge. There was no significant difference with respect to bleeding when comparing the transscleral group with the transpupillary group.
In 2/37 eyes (5%) an unfavourable outcome was observed. The first case concerned a child with bilateral threshold ROP. In addition, a severe anterior segment ischaemia was present in the right eye. After transpupillary laser photocoagulation (2000 burns in total in both eyes) anterior and posterior chamber haemorrhage occurred in the right eye. At the same time intracerebral haemorrhage, hydrocephalus with multiple shunt operations and revisions, pulmonal artery stenosis, a bronchopulmonal dysplasy, cerebral cramps and secondary sepsis resulted in a life-threatening situation. After resorption of the blood, cataract formation was observed. After cataract surgery the aphakic eye was corrected with a soft contact lens for a hypermetropia of +34 D. The left eye responded to laser therapy. The refractive error at 20 months was +5.5 sph −2.5 cyl × 0. The right eye was found to be amblyopic at the last follow-up examination.
The second case with an unfavourable outcome concerned a girl born at 24 weeks gestational age with a birth weight of 560 g. The infant could not be operated at the time that threshold ROP was diagnosed because of life-threatening lung and intestinal problems. Retinopathy of prematurity had already progressed to stage IV by the time an ophthalmic intervention was possible. Transscleral laser treatment was performed on both eyes. When her left eye progressed to stage IV B, an encircling band was placed. However, the encircling procedure was unable to halt progression to stage V disease and a vitrectomy had to be initiated.
Postoperative refraction was evaluated by retinoscopy twice during follow-up. Thirteen infants (68%) were followed up at both time-points, while two patients were measured at either the first or second follow-up visits (Table 2). The spherical equivalents of the refractive errors for both time-points are shown in Fig. 1. The majority of values ranged between +4 D and −2 D. At the time of the first retinoscopy 14 patients (27 eyes) were included. The infants' mean age was 23 ± 12 months (range 5−44 months). The mean spherical equivalent per refracted eye was −1.0 ± 4 D. One eye was excluded from analyses of means because of hypermetropia of +34 D due to aphakia after cataract extraction. The eye was corrected with a soft contact lens. The majority of the eyes (85%) were hyperopic (range 0.75 − 6.0 D). One eye was emmetropic and only three eyes (11%) were myopic, with a mean of − 6.7 D. The maximum values for astigmatism were measured at 1.5 D (Table 3).
Table 2. Follow-up of the refractive error in spherical equivalents: 15 patients were examined twice while two patients were refracted at either the first or the second retinoscopy visit.
|1||+ 1.5||+ 1.5||5||+ 1.5||+ 1.5||24|
|2||+ 34.0|| ||7||+ 34.0||+ 4.25||20|
|3||+ 2.0||+ 3.0||12||+ 2.5||+ 3.0||20|
|4||+ 0.75||+ 1.0||12||+ 1.0||+ 1.25||36|
|5|| || || ||− 9.5||− 1.25||41|
|6||− 7.0||0||18|| || || |
|7||− 8.0||− 5.0||12||− 12.0||− 5.25||39|
|8||+ 1.75||+ 1.25||24||+ 1.25||+ 1.0||40|
|9||+ 3.5||+ 3.5||26||+ 3.5||+ 3.5||46|
|10||+ 0.5||+ 0.5||24||+ 2.0||+ 2.0||50|
|11||+ 6.0||+ 6.0||23||+ 7.0||+ 7.25||53|
|12||+ 0.75||+ 0.75||21||+ 2.0||+ 2.0||54|
|13||+ 2.25||+ 1.75||40||+ 2.25||+ 1.75||62|
|14||+ 4.0||+ 4.0||40||+ 5.75||+ 6.0||65|
|15||+ 1.0||+ 1.0||44||+ 1.25||+ 1.0||60|
Figure 1. The distribution of the refractive errors of all eyes is shown in the graph. For the 13 patients who were examined at both visits the values of the eyes are linked with one another in order to demonstrate the refractive shift of each eye during follow-up. In five eyes only a single refractive error was obtained; these eyes are illustrated by non-linked single symbols. In general the majority of the refractive errors ranged between −2 D and +4 D.
Download figure to PowerPoint
Table 3. First retinoscopy: refractive results after laser treatment in 27 eyes (mean age 23 ± 12 months).
|Patient||Right eye||Left eye|
|1||+ 1.5 sph||+ 1.5 sph|
|2||+ 1.0 sph − 0.5 cyl × 0°||+ 1.0 sph|
|3||+ 34.0 sph|| |
|4||+ 2.0 sph||+ 3.0 sph|
|5||− 7.0 sph||0.0 sph|
|6||− 8.0 sph||− 5.0 sph|
|7||+ 6.5 sph − 1.0 cyl × 180°||+ 6.5 sph − 1.0 cyl × 180°|
|8||+ 1.0 sph − 0.5 cyl × 0°||+ 1.0 sph − 0.5 cyl × 0°|
|9||+ 3.0 sph − 1.5 cyl × 0°||+ 2.5 sph − 1.5 cyl × 170°|
|10||+ 4.5 sph − 1.0 cyl × 0°||+ 4.5 sph − 1.0 cyl × 0°|
|11||+ 1.75 sph||+ 1.0 sph − 1.5 cyl × 180°|
|12||+ 3.5 sph||+ 3.5 sph|
|13||+ 1.0 sph||+ 1.0 sph|
|14||+ 0.5 sph||+ 0.5 sph|
Fourteen children (28 eyes) with a mean age of 45 ± 14 months (range 20−65 months) were investigated at the time of the second retinoscopy. The mean spherical equivalent per refracted eye was +1.1 ±3.3 D. The values ranged between −8 D and +6 D, except for the aphakic eye that was excluded from analyses. In the other eyes hypermetropia was present in 24/28 eyes (86%). Except for one eye with 3 cyl, astigmatism was low (Table 4).
Table 4. Second retinoscopy: refractive results after laser treatment in 28 eyes (mean age 45 ± 14 months).
|Patient||Right eye||Left eye|
|1||+ 34 sph (aphakic)||+ 5.5 − 2.5 × 0°|
|2||+ 1.25 sph − 0.5 cyl × 0°||+ 1.5 sph − 0.5 cyl × 0°|
|3||+ 2.5 sph||+ 3.0 sph|
|4||− 12.0 sph||− 5.25 sph|
|5||− 9.5 sph||− 1.5 sph − 1.0 cyl × 0°|
|6||+ 4.5 sph − 1.25 cyl × 0°||+ 4.0 sph − 0.75 cyl × 0°|
|7||+ 2.0 sph||+ 2.0 sph|
|8||+ 7.75 sph − 1.25 cyl × 167°||+ 8.75 sph − 3.0 cyl × 179°|
|9||+ 1.5 sph − 0.5 cyl × 0°||+ 1.5 sph − 0.75 cyl × 170°|
|10||+ 1.5 sph||+ 1.5 sph|
|11||+ 3.0 sph − 1.5 cyl × 0°||+ 2.5 sph − 1.5 cyl × 0°|
|12||+ 2.0 sph||+ 2.0 sph|
|13||+ 6.5 sph − 1.5 cyl × 0°||+ 6.75 sph − 1.5 cyl × 10°|
|14||+ 1.5 sph − 0.5 cyl × 180°||+ 1.0 sph|
Three infants (17%) developed an esotropia and muscle surgery has been performed in one of these patients so far.
- Top of page
- Materials and Methods
Besides the immediate short-term results it seems that laser therapy may also be beneficial in respect to longterm functional and visual outcome. The incidence of myopia after therapy for ROP has been shown to range between 16% and 50%. Eyes treated with cryotherapy have been found to be significantly more myopic than those treated with laser photocoagulation (Iverson et al. 1991; McNamara et al. 1992; Hunter & Repka 1993; Quinn et al. 1998). Not only were the cryotherapy-treated eyes more myopic in the early post-treatment period, but they also experienced more myopic shift compared to the laser-treated eyes. This trend of increasing myopia has been previously reported in both laser-treated and cryotherapy-treated eyes up until the age of 6 years (Kent et al. 2000). However, Quinn et al. (2001) published data indicating that after 12 months no significant change in the refractive error between control and ROP treated eyes was detectable. Our results support these findings. We found no significant difference between the refractive results of 13 patients treated by laser 23 ± 12 months after treatment and the results 45 ± 14 months after treatment (Table 2).
Recently, it has been suggested that the power of the lens may be responsible for myopia in preterm infants and that the incidence of myopia is lower after laser therapy than after cryotherapy (Connolly et al. 2002). The mean lens power for age-matched patients was shown to be +18.9 D, while it was +22.7 D in the laser-treated eyes and +29.9 D in the cryotherapy- treated eyes. In contrast, both the axial length and the posterior segment length were found to be shorter in the cryotherapy-treated eyes, while the anterior chamber was shallower and the lens was thicker.
Late onset or persistent treatment complications after laser therapy are rarely described in the literature. Complications of the anterior segment can include burns to the cornea, iris and tunica vasculosa lentis with induction of cataract (Irvine et al. 1990; McNamara et al. 1991; Landers et al. 1992). The incidence rates of cataract development, corneal changes and a coagulation of the tunica vasculosa lentis seem to be low when using the transscleral technique (Seiberth et al. 1997). In our study none of these complications occurred except for cataract formation in one eye (3%) that had received transpupillary laser coagulation. In this eye with severe anterior ischaemia, bleeding into the anterior and posterior chambers occurred after laser application. Resorption of these haemorrhages was seen within 6 weeks. Apart from this severe bleeding, only minor spot bleeding at the ridge was observed in 8% of our patients. These haemorrhages were resorbed spontaneously. Postoperative intraocular haemorrhages located centrally from the ridge were detected in another five eyes (14%). There was no significant difference between the transscleral and transpupillary laser techniques with respect to the occurrence of haemorrhages. This incidence of haemorrhage is consistent with the findings of Seiberth et al. (1997), who observed bleeding in 36% of transsclerally treated eyes versus 20% of transpupillary treated eyes.
Three infants developed an esotropia and in one eye out of the 37 the retina detached despite laser photocoagulation.
We conclude that laser treatment showed favourable results in our study, inducing regression of threshold ROP in 97% of all treated eyes. The eye that did not respond to laser treatment was diagnosed at stage III +0. Due to severe life-threatening lung and intestinal problems the laser treatment had to be postponed and stage IV had already developed by the time laser photocoagulation could be carried out. This is why we think this eye continued with progression despite laser therapy, encircling band and vitrectomy.
In all, 87% of our patients received transscleral laser treatment. Initially, this was only carried out when the size of the pupil dilated inadequately. However, as the study progressed we began to prefer the transscleral technique. In 1999 the transscleral laser treatment had been reported as being a technically straightforward and effective alternative to cryotherapy (Davis et al. 1999). This technique allows the surgeon to control the intensity of each single laser burn in contrast to the transpupillary mode, where the burns may turn out to be too intense when there are sudden changes in the thickness or pigmentation of the retina. Furthermore, the ‘pop effects’ that have been described with transpupillary administration are highly improbable with transscleral treatment (Seiberth et al. 1997). In transscleral administration the intraocular distance to the target tissue is minimal compared to the transpupillary technique, where the laser beam has to traverse the anterior and posterior segments of the eye before reaching the retina. Concurrent to Seiberth et al. (1999), we hypothesize that not having to penetrate the whole eye may be advantageous in general. In addition, it may result in decreasing the incidence of cataract formation. In our case series only one cataract occurred and this was in an eye that had received transpupillary laser administration. The reduction of the incidence of cataract formation has already been described by Seiberth et al. (1997), who also reported that transscleral laser causes more trauma because of the necessity to incise the conjunctiva in cases of ROP in central zones II and I. Our findings differ from these results insofar that we only had to perform a conjunctival incision in 14% of our cases. Furthermore, transscleral laser photocoagulation was in general less time-consuming than transpupillary laser coagulation.
The lack of controls and the lack of randomization limit the results of our study. Nevertheless our findings are encouraging with respect to the efficacy and safety of laser photocoagulation as a valuable alternative to cryotherapy. Transscleral laser photocoagulation appeared to be particularly advantageous, less traumatic and less time- consuming, providing control of the intensity of each single laser burn and thus allowing for the administration of a perfect dosage throughout the whole procedure. Furthermore, we are convinced that transscleral laser photocoagulation can also be performed under local anaesthesia and additional sedation in most cases except for central ROP eyes that require incision of the conjunctiva. The analyses of the refractive values obtained so far indicate that there might be a decrease in myopia after laser therapy versus cryotherapy. Randomized, controlled, prospective, clinical trials are necessary to further validate these results.