Introduction

Retinopathy of prematurity (ROP) is a disease of premature infants, characterized by retinal ischemia, fibrovascular proliferation, retinal and vitreoretinal neovascularization, and progressive vitreoretinal traction. The incidences of childhood blindness due to ROP range from 3 to 10% worldwide. Depending on the developments in neonatal care, ROP is becoming increasingly important, especially in developing countries [1,2,3,4]. Following early treatment for ROP (ETROP) study, laser photocoagulation (LPC) was accepted as the standard treatment for infants in type 1 ROP [5, 6]. However, it has been shown that laser therapy causes a number of unfavorable anatomic and functional outcomes, especially in cases with zone I and posterior zone II ROP [7, 8].

On the other hand, overexpression of vascular endothelial growth factor (VEGF), which causes abnormal vascular proliferation in premature infants, is known to play an important role in the pathogenesis of ROP. This situation has increased the intravitreal use of anti-VEGF drugs in the treatment of ROP and hence the results in rapid resolution of the plus disease with regression of ROP today [9,10,11].

Long-term studies have shown that children born prematurely have much more refractive errors than those born at full term [12, 13]. On the contrary, in severe ROPs requiring treatment, myopia and other refractive errors are more common than premature infants, and this increases the risk of amblyopia [14,15,16].

The aim of the present study was to evaluate 1-year refraction values in preterm infants to whom IVB and LPC therapy were applied.

Materials and methods

The Ethical Review Committee authorized this retrospective study (2022/01), which followed the standards of the Declaration of Helsinki for research involving human subjects. The study included preterm infants who received LPC and IVB therapy for ROP.

According to the type of treatment, the patients who underwent LPC were classified as Group 1, and the patients who received IVB were classified as Group 2. Infants who received additional and combined treatment and those who underwent surgical treatment were excluded. The infants’ gestational age (GA) at birth, birth weight (BW), ROP stages and zone of the patients, treatment options, and corrected first age refraction values were recorded. All patients' spherical and cylinder power as well as spherical equivalent (SE) were recorded as diopters (D). ROP screening was performed on infants with a birth weight of  ≤ 1500 g and a gestational age of  ≤ 132 weeks and selected infants considered at risk by the neonatologist.

ROP status of the patients was determined according to the severity of the disease according to the criteria of the International Classification of ROP, third edition (ICROP-3) [11].

The cases that met the criteria for type I ROP (stage 2 or 3 in zone II with plus disease, stage 3 in zone I with or without plus disease, or stage 1 or 2 disease in zone I with plus disease, as defined by the ETROP study) received LPC therapy [5] and the cases diagnosed with A-ROP (rapid development of pathological neovascularization and severe plus disease without typical progression to ROP stages) received IVB therapy.

LPC was indicated for preterm infants with type 1 ROP and applied using 810-nm transpupillary diode laser (OcuLight® SL, Iridex, USA) to all avascular areas at half-spot size intervals under remifentanil sedo-analgesia. IVB treatment was indicated for cases with A-ROP, and 0.625 mg/0.025 mL IVB (Altuzan® 100 mg/dl, Roche, Sweden) was injected into the diseased eye using a sterile 32-gauge needle, at 1.5 mm from the limbus with aseptic condition using topical anesthesia. Follow-up care was conducted on postoperative day 1 and continued on a weekly basis for the first month and once every 2 weeks after 2 months until complete regression of ROP and retinal vascularization reached the temporal ora serrata.

After these examinations, all infants were scheduled to be called for the 6th month and 1-year controls and screened for refractive errors, anterior and posterior segment development, and retinal status. Refractive errors were evaluated using cycloplegic retinoscopy and automated refractometry (Welch Allyn SureSight Autorefractor, USA), following dilatation with two drops of 1% cyclopentolate hydrochloride (Sikloplejin, Abdi İbrahim, Turkey).

The SPSS 25.0 program package was used for statistical analysis. Data was presented as frequencies and percentages or as the mean ± SD. Chi-square test and Fisher’s exact test were used for categorical variables. Normal distribution fitting was checked with Kolmogorov–Smirnov test. The differences between the means were carried out by using the t-test for normally distributed data and the Mann–Whitney U test for the data that did not conform to the normal distribution. Univariable and multivariable linear regression analyses were used to investigate the effect of gestational age and birth weight on refractive results. P values of 0.05 or lower were considered to indicate statistical significance.

Results

The study included 86 eyes of 45 infants with a mean ± standard deviation gestational birth age of 26.5 ± 2.1 weeks (range: 23–31 weeks) and a mean birth weight of 904 ± 223 g (range: 590–1470 g). There were 23 (51.1%) male patients and 22 (48.9%) female patients. There was no significant difference between the groups considering gestational age (GA), birth weight (BW), and gender distribution (P = 0.349 for GA, P = 0.109 for BW, and P = 0.673 for gender).

LPC therapy was performed on 59 eyes of 30 infants diagnosed with type I ROP (Group I), and IVB therapy was performed on 27 eyes of 15 infants diagnosed with A-ROP (Group II). Of the 30 infants treated with LPC, 58 eyes of 29 infants had zone II ROP and one eye of one infant had zone III ROP. Of the 15 infants treated with IVB, 14 eyes of eight infants had zone I ROP and 13 eyes of seven infants had zone II ROP.

Treatments were administered up to a mean PMA of 36.0 ± 2.4 (range: 31–42 weeks) and 35.3 ± 2.6 (range: 32–40 weeks) weeks in Group I and Group II, respectively (P = 0.32). No complications, such as iatrogenic cataract, intraocular inflammation, retinal detachment, and vitreous hemorrhage, were observed due to the treatment modality applied to the patients.

In the follow-up examinations, 1-year spherical, cylindrical, and SE values were 0.1 ± 2.2 D, − 1.2 ± 0.9 D and − 0.5 ± 2.0 D in Group I and 1.3 ± 1.7 D, − 1.1 ± 0.8 D and 0.8 ± 1.7 D in Group II, respectively (P = 0.018 for spherical, P = 0.772 for cylindrical, and P = 0.009 for SE). Myopic refraction was significantly greater in Group I than in Group II (P = 0.018). Demographic data of infants and corrected first age refraction values of infants are presented in Table 1.

Table 1 Demographic data and corrected first age refraction values of infants
Full size table

The mean spherical power and spherical equivalent were significantly different between the two groups for zone II (p = 0.005 and p = 0.002). However, no significant difference was found in the mean cylindrical power (p = 0.648). Refraction values of infants according to ROP zones in Group 1 and Group 2 patients are presented in Table 2.

Table 2 Refraction values according to ROP zones in Group 1 and Group 2 patients
Full size table

In eyes treated with IVB, mean spherical power and spherical equivalent were found to be significantly lower in zone I than in zone II (p = 0.04 and p = 0.034). In addition, when we compared the effects of treatment stages in the development of myopia in the eyes treated with LPC, we found that the SE values in Stage 2 and Stage 3 patients were 0.1 ± 1.2 D and − 1.1 ± 2.5 D. Infants with Stage 3 ROP were found to be significantly more myopic than infants in Stage 2 ROP (p = 0.03).

The distribution of refraction results grouped as very high myopia (≤  − 8 D) to high hyperopia (≥ + 4 D) is shown in Fig. 1. Emmetropia (− 1.0 to + 1.0 D) and low hyperopia (≥ + 1.0 to + 4.0 D) were significantly higher in eyes treated with IVB (p = 0.008 and p = 0.001). Furthermore, linear regression analysis revealed no significant correlation between mean SE of gestational age and birth weight (p = 0.25 and p = 0.54).

Fig. 1
figure 1

The distribution of refraction results of the groups according to spherical equivalence values. Results grouped as very high myopia (≤-8.0 D), high myopia (−8.0 to −5.0 D), low myopia (−5.0 to −1.0 D), emmetropia (−1.0 to +1.0 D), low hyperopia (≥+1.0 to +4.0 D), and high hyperopia (≥+4.0 D)

Full size image

Discussion

The present study observed lower SE values following laser photocoagulation therapy (− 0.5 ± 2.0 D) compared to IVB therapy (0.8 ± 1.7 D). The mean spherical power and spherical equivalent were found to be significantly lower in zone I than in zone II in the group treated with IVB. In addition, when the effects of treatment stages in the development of myopia in patients treated with laser were compared, eyes with Stage 3 ROP were found to be significantly more myopic than eyes with Stage 2 ROP.

In recent years, a series of studies evaluating refractive outcomes following IVB and laser therapy have shown that IVB is associated with a lower rate of refractive errors in ROP compared to laser ablation [17,18,19,20]. Similarly, Harder et al. [18] reported that the mean SE in laser-treated infants was  − 4.4 ± 1.7 D, which is much more myopic than  − 1.04 ± 4.24 D in the IVB-treated preterm group.

By contrast, some studies reported that refractive errors did not differ significantly between anti-VEGF drug and laser treatment [21,22,23]. Although the cause of myopization in children treated with ROP has not been fully understood to date, Fielder and Quinn suggested that ROP treatment-related myopia may develop due to the physical restriction of anterior segment development of the damaged peripheral retina caused by laser therapy [24]. On the contrary, it has been suggested that IVB treatment blocks VEGF that is already produced and present in the vitreous [20], and after treatment, retinal development may maintain the local growth factor expression, providing the continuation of signaling pathways, and thus allowing the anterior segment to develop normally, leading to a decrease in myopia [17].

Hwang et al. [19] reported mean cylindrical values of 1.6 ± 1.5 D and 1.0 ± 0.8 D in eyes treated with laser and IVB, respectively. No significant difference was found between the groups in terms of cylindrical values in eyes with zone I and zone II ROP (for zone I p = 0.13; for zone II p = 0.19). On the other hand, Harder et al. [18] reported that refractive astigmatism was significantly lower in the single-dose IVB group compared to the conventional retinal laser photocoagulation group (−1.00 ± 1.04 D for IVB, 1.82 ± 1.41 D for laser, p = 0.03). However, in this study, no significant difference was found between LPC therapy (− 1.2 ± 0.9 D) and IVB therapy (− 1.1 ± 0.8 D) in terms of astigmatism. The difference in the distribution of patients in zone I and zone II between treatment groups and the absence of patients in zone I in patients treated with LPC may have caused the indifferences between groups in astigmatism.

In the BEAT-ROP clinical trial comparing the refractive results between IVB and laser therapy, it was reported that eyes with zone I ROP (− 1.51 D for IVB therapy and  − 8.44 D for laser therapy) had a higher degree of myopia than eyes with zone II ROP (− 0.58 D and  − 5.83 D for IVB therapy), but no difference was found for the IVB arm or the laser arm. Therefore, the authors thought that the severity of the disease was not the main factor influencing the severity of myopia [17]. Hwang et al. [19] evaluated refractive errors according to the zones, and it was reported that eyes with zone I ROP were more myopic than eyes with zone II ROP, especially in the IVB group. Similarly, Gunay et al. [25] reported that eyes with zone I ROP were more myopic than eyes with zone II ROP in both the IVB and laser groups. In this study, when the effects of treatment zones in the development of myopia were compared, in patients treated with IVB, eyes with zone I ROP were found to be significantly more myopic than eyes with zone II ROP (p = 0.04).

Davitt BV et al. [26] suggested that the prevalence of myopia and high myopia increased in association with the presence of plus disease and increasing ROP stage (severity).

On the other hand, Pennefather et al. [27] showed that advanced ROP (stage 3 or worse) is associated with the development of >  − 0.25 D myopia, but in lesser ROP, this risk is no greater than it is in the general population.

Consistent with the literature, infants with Stage 3 ROP were found to be significantly more myopic than infants in Stage 2 ROP (p = 0.03) in our study.

This study includes limitations such as retrospective design, small sample size, lack of ocular biometry, and follow-up time. In this study, only refractive error was evaluated. Therefore, there may also be a need for other biometric measures to evaluate the myopia associated with ROP [28]. The follow-up period of approximately 1 year did not allow conclusions be drawn about the long-term effects and side effects of the treatment.

In conclusion, this study supports that even 0.625 mg intravitreal bevacizumab for ROP causes less myopia compared to laser photocoagulation and that myopia is higher in patients with advanced ROP. In future, larger and longer-term studies are needed to elucidate the effect of laser photocoagulation or IVB therapy on refractive error in eyes with ROP.