Purpose The aim of this study was to compare all retinal layers' thickness in full-term and preterm children without retinopathy of prematurity (ROP). Methods Cross-sectional study including two groups of patients: group 1 children with history of preterm gestation without ROP (gestational age < 37 weeks) and group 2 healthy children with history of full-term gestation. All subjects underwent an ophthalmic examination including spectral domain-optical coherence tomography. After automatic retinal segmentation, each retinal layer thickness (eight separate layers and overall thickness) was calculated in all nine Early Treatment Diabetic Retinopathy Study areas. Demographic, systemic, gestational, and birth data were collected. Generalized additive regression models were used to analyze the data. Results Fifty-one children (51 eyes) were recruited, 19 full-term and 32 preterm children, mean age at ophthalmic examination of 10.58 (4.21) and 14.13 (3.16), respectively. In multivariable analysis, the preterm group's retinal thickness was significantly decreased in total retina nasal outer sector, ganglion cell layer (GCL), and inner plexiform layer (IPL), specifically GCL temporal outer (p = 0.010), GCL superior outer (p = 0.009), IPL temporal outer (p = 0.022), and IPL superior outer (p = 0.004), when compared with full-term group. From the variables compared only with birth head circumference that influenced the models, a non-linear association was identified and consequently modeled with splines through a generalized additive model. Conclusion This study suggests that preterm children without ROP have structural retinal alterations, mostly in GCL and IPL in outer areas of the macula. Therefore, it is crucial to question gestational history since these retinal changes may be found later in life leading to useless investigation.