TON is a type of optic nerve injury that occurs as a result of trauma. RGCs degeneration due to axonal damage may also occur in patients with TON. In addition, as TON causes loss of RGCs and their axons, the optic nerve damage due to TON can affect the RNFL and GCIPL. Given that RGC synapses are located in the inner reticular layer, alterations in this layer are expected after optic nerve injury [11]. Timely diagnosis and appropriate management of TON rely on a comprehensive understanding of its characteristics and pathophysiology, which is important for determining diagnosis and treatment. Although TON may not exhibit abnormal retinal findings during early phases, changes over time due to nerve damage can be monitored using OCT [12, 13].

Some studies have reported retinal layer thickness to investigate the decrease in retinal activity after optic nerve injury. They reported up to a 40% reduction in the thickness of the RNFL in five TON patients [8]. Others investigated the distinct differences in RNFL and GCIPL thickness between the traumatic eyes and fellow eyes in TON patients whose onset of blunt trauma was valuable from one month to 55 years [9]. However, most studies had small sample sizes and did not investigate longitudinal changes in retinal morphological features and visual function in TON patients. In addition, most studies have not divided TON according to the mechanism of damage. Therefore, this study was conducted with a larger sample size and compared the two types of blunt and surgical injuries to evaluate the clinical course of TON patients.

Functional markers such as visual acuity and VFI showed significant reductions in the traumatic eye compared with the fellow eye from the first post-injury visit and over a 1-year follow-up period in both indirect and direct TON patients. Regarding the morphological findings with RNFL and GCIPL thickness changes, RNFL thickness showed an early loss in blunt TON compared to the surgical TON. On the other hand, GCIPL thickness was observed to decrease simultaneously in both blunt and surgical TON at 1 month after the first visit. Thickness change in GCIPL might be a more sensitive parameter to predict retinal adversity in TON despite the cause of optic nerve damage. It also suggests that RNFL thickness may be a useful marker to determine retinal affliction in blunt TON.

Ischemia probably plays a pivotal role in both the initial neuronal edema and the subsequent neuronal cell death and axonal atrophy seen in different stages of TON [14]. Ustymowicz. et al. documented a significant reduction in blood flow within the central retinal artery of affected eyes in TON patients [15]. Additionally, Shi et al. observed that changes in the parameters of the central retinal artery, which primarily supplies blood to the optic nerve and inner retina, were particularly pronounced among the vasculature parameters studied [16]. Furthermore, surgical TON may be more susceptible to ischemic conditions compared to blunt TON, potentially due to additional neurological stresses. These stresses include inflammatory mediator triggering macrophage and polymorphonuclear cell infiltration, resulting in the release of enzymes and free radicals that can further damage axons [17, 18].

The time of post-traumatic retinal morphological changes should be particularly noted in relation to the time of optic atrophy. Optic atrophy is a well-known consequence of TON and is characterized by a gradual pale and atrophic optic disc that occurs 3–5 weeks after trauma [19]. However, we detected a tendency of RNFL thickness to decrease and a marked reduction in GCIPL thickness in TON patients. This finding suggests that the morphological changes in GCIPL may precede the changes in RNFL in early TON and that it may occur before optic atrophy. This is an important observation as it suggests that GCIPL thickness may be a more sensitive parameter for predicting retinal injury in TON, regardless of the cause of optic nerve damage. Furthermore, RNFL thickness may be a useful marker for determining retinal damage in blunt TON. Overall, this finding highlights the importance of monitoring both RNFL and GCIPL thickness in TON patients to predict and distinguish the morphological destruction and function deterioration according to the types of TON.

Several limitations should be considered when interpreting the findings of this study. Firstly, the retrospective design of the study may have introduced inherent biases and limitations, such as selection bias and the inability to establish causality. Moreover, the relatively small sample size, encompassing only 18 patients followed up for 1 year, might restrict the generalizability of the findings. The lack of randomization and potential confounding variables among the blunt and surgical TON groups could also affect the comparability of the outcomes. Additionally, the handling of missing data using regression analysis for imputation may introduce potential biases and affect the robustness of the results.

Despite the aforementioned limitations, this study offers several strengths that contribute to the understanding of TON. The inclusion of both blunt and surgical TON patients, along with a comprehensive evaluation of retinal morphological features and visual function, provides valuable insights into the clinical course of TON patients. The longitudinal evaluation of retinal layer thickness and functional markers over a 1-year follow-up period supplements existing literature by capturing the dynamic changes in TON progression. Furthermore, the focus on distinct variances in retinal layer thickness between affected and unaffected eyes, as well as the comparison of blunt and surgical injuries, enhances our understanding of the morphological and functional changes associated with different mechanisms of optic nerve injury.

In conclusion, regardless of the limitation as a retrospective design, we could discover the significant thinning of RNFL and GCIPL in both blunt and surgical TON eyes, with remarkable reduction of GCIPL in early phase. Therefore, analyzing each retinal layer thickness using OCT as well as visual function would be helpful to understand clinical course of each TON and predict the morphological destruction and function deterioration in TON. Moreover, evaluating not only the RNFL thickness but also the GCIPL thickness in traumatic eyes could be useful to predict and distinguish the morphological destruction and function deterioration according the types of TON. Further studies with larger sample size and longer follow-up periods are warranted to validate the findings.