In our study, the refractive development patterns of premature infants without ROP were similar to those of full-term infants, i.e., the refractive status among 1-18-months-old children reflected hyperopia that became emmetropia with age, and this was consistent with previous studies[12, 13]. As expected, the MSE of premature infants was lower than full-term infants in each age group (P = 0.749), which the finding have been also reported[14]. In addition, our data showed that neonatal hyperopia rapidly decrease and turn into emmetropia, the MSE in premature infants non-ROP after one year of age were significantly lower than in full-term infants (P = 0.022), and that the premature infants manifested more significant emmetropia than full-term infants (62.7% vs. 38.3%, P = 0.037). Tian et al.[15] has reoprted that when the full-term children reached emmetropia among 0–6 years old, the refractive status of most premature infants of the same age had developed into myopia. This effect may be caused by the eyes of premature infants possess a spherical lens, a shallow anterior chamber, and greater corneal curvature[16–19]. Al et al.[20] compared the eyeball shape of premature infants non-ROP and full-term infants, and found that small pathologic changes and retinal traction occurred around the retina or that early optic nerve ischemia was present, even in the absence of significant ROP; this exerted detrimental effects on refractive development or emmetropization in their premature infants. We therefore speculate that premature infants without ROP still possess the possibility of “catch-up-growth” in refractive development after one year of age, but more participants are needed to discuss the refractive status of premature infants at different postnatal ages in a longitudinal study.
Premature departure from the intrauterine environment may affect ocular development, and that there exists an imbalance in the development of various refractive parameters of the eyeball. Postnatal diseases may also accelerated emmetropization during a critical phase of vision development and thus generate a higher prevalence of refractive error, but this is not solely limited to ROP infants[21]. In this research, we estimated the diffrences of refravtive error between premature infants without ROP and full-term infants based on PSM that resembles randomized experiments. The results of the present study revealed that the prevalences of hyperopia and astigmatism gradually decrease with increase in age while the prevalence of myopia did not change significantly of premature infants. According to Snir[14], premature infants with a corrected gestational age (CGA) of 40–44 weeks showed a tendency toward myopia, whereas, although the CGA of enrolled premature infants was 40 weeks in the present study, preterm infants showed no greater tendency to myopia than full-term infants. We suspect that the reasons may be that premature birth only has an early efect on the diopter values of preterm infants, and the growth and development of the premature have reached the full-term state when we recruited them. Fielder[22] proposed a hypothesis that myopia was probably the normal refractive state in infants before full-term, but in this reserch, we missed that stage.
Similar to previous studies[23], current study showed the prevalence of astigmatismt was highy predictive of significant in premature infants non-ROP (56.1% vs. 47.4%, P = 0.040). A rencent[24], large, cross-sectional study in China has shown that the prevalence of astigmatism was highest in the ROP group, followed by the non-ROP group and that full-term infants manifested the lowest prevalence. Deng[25]compared ocular morphology between premature infants non-ROP and mature infants at 6-12months and found that shorter axial length, shallower anterior chamber, and higher curvature of the cornea in the preterm infants groups. The study of Chinese children’s results might explain the reason of high prevalence of astigmatism and myopia in preterm infants in childhood. In other words, premature delivery is possble a risk factor for lead to blocked development of the anterior segment of the eye, which affected the process of orthokeratology and refractive status [26].
In addtion, we ascertained that the astigmatism axis distribution of premature infants without ROP was similar to that of full-term infants with WTR occupying a dominant position, However, the proportion of OBL astigmatism of premature infants non-ROP was far higher than that mature infants (76.5% vs. 23.5%, P = 0.026). In a follow-up study, Danish children aged 3–9 years, OBL astigmatism was related to myopic progression and exerted a greater impact on pediatric vision function[27], and we herein found that astigmatism was particularly prominent in premature infants at 1–18 months as persistent high astigmatism was present (particularly OBL astigmatism). This effect may block the retina from forming images clearly and disrupt the emmetropization process, resulting in a risk of meridional amblyopia or the induction of myopia[28, 29]. Recognition of high astigmatism and/or astigmatism at oblique axis in these premature infants is important, the findings are beneficial for refractive error tracking and vision correction for preterm infants, prompt treatment helps these children achieve better visual acutiy.
The early refractive status of premature infants is affected by many factors, including GA, BW, corneal curvature, and axial length. Wood[30] examined the significant differences in ocular optical components between premature infants and full-term infants and attempted to resolve the disparities in refractive status from the perspective of eyeball biology. These authors also found that refractive status in premature infants was related to physical parameters at birth. In 2009, Varghese[31] enrolled 599 one-week-old infants for ophthalmologic examinations, and their linear regression analysis determined that the refractive status of premature infants was positively correlated with GA and BW, but unrelated to BL and head circumference at birth. We herein determined the refractive distribution of premature infants with different GAs, BWs, and BLs. Moreover, in our study, statistically significant differences in MSE between premature infants with different BW, demonstrating that MSE was lower when BW was higher (P = 0.049). Further linear regression analysis was conducted on the two factors, and our results revealed that BW was negatively correlated with MSE (R2 = 26.01%, P < 0.001), i.e., as birth weight increased, MSE decreased, which was consistent with the data from previous studies[32, 33]. Even though BL was negatively correlated with diopter, the fit was low (R2 = 3.54%). Therefore, we deduce that BW is one of the most reliable physical parameters at birth, and that it may affect refractive status of infants aged 1-18months .
In addition, There was no evidence that significant correlation between GA and MSE in premature infants without ROP(P = 0.227), which was in contrast with the results of aforementioned studies where the various biological measurements of premature infants were significantly lower than in full-term infants when GA was lower[34, 35]. The increase in axial length in premature infants also did not compensate for the diminution in corneal curvature and lens refraction and tended to cause myopia and astigmatism[36–39]. The disparities between our results and previously in the literature may be due to the fact that our investigation did not include premature infants with ROP, and that the GA and BW thresholds we applied for grouping and the age at examination were different. We hypothesize that although GA is related to early diopter in premature infants after birth, this relationship does not continue with increasing age, but still needs to be validated by more experiments.
Potential limitations of our study should be noted. Even though we used PSM analysis, inconsistencies in the baseline and bias could still be possible. we collect information from premature infants without ROP, but the babyies may be with slight or regressed ROP not noticed by doctors. In addition, more studies with large sample size are needed to investigate the influence of prematurity on the basis of corrected GA, BW and BL.