Corneal transplantation is one of the important therapeutic options to restore vision in some corneal diseases. Although the survival rate of corneal graft is still affected by serious causes, the initial success of corneal graft is close to 90%. In recent decades, the frequency of corneal regrafts has increased. This can probably be attributed to the greater availability of donor tissue resulting from increasing eye-banking activities and the growth of the number of corneal transplantations. [9–12]
In this study, the mean age of the regraft recipients was (50.9 + 21.8 years) similar to another study from Brazil and Canada. [13, 15] Weisbrod et al showed that the mean age of their patients that underwent regraft surgery was 5 years older than those undergoing corneal graft for the first time. More corneal graft failures at younger ages could have been either because of a higher proportion of complicated cases [e.g., corneal dystrophy, ocular trauma] or poor compliance with postoperative care among younger ages. [13, 15] The sex ratio in our study was similar to the Indian study where 68% of the patients were men [15, 16], but not follow the others studies realized in developed countries. [13, 14]
Our study showed that the rate of corneal regraft in all transplanted corneas during ten years was 8.55% which is similar to the corneal graft failure rate (5–30%) in other studies [7, 8, 17, 18]. In this study, DALK was the most successful type of grafting with a regraft rate of 1.08% followed by PK (8.32%) and DSAEK (14.98%). Most probably reasons for the acceptable rate of regrafting in DALK cases were younger age at the time of surgery, the nature of the main disease (keratoconus in most of our DALK cases), and retaining recipients’ endothelium. Considering the time interval between the primary graft and regrafting (1.5–6 months), the failure was most probably caused by technical problems [19]. Considering the strict DSAEK technique and less familiarity of our cornea surgeons with the DSAEK technique at the beginning years of DSAEK introduction, the success rate was still very acceptable in our tertiary center. Other reasons for the higher rate of re-DSAEK in our center can be attributed to the special socio-climatic conditions that prone the most patient to ocular surface complications, and due to the higher prevalence of ocular surface problems in PK patients, it has been attempted in recent years even in most cases of PK failure due to endothelial dysfunction, primary regraft surgery is DSAEK.
In our study, in contrast to previous studies, infectious causes (corneal ulcer) of the grafted cornea were the main cause of regrafting [4, 5, 17]; however, following most studies, endothelial dysfunction had a more significant role in corneal regraft [5, 6, 17, 18]. It can be due to climacteric differences or social-cultural conditions of most of our cases who are not able to follow the recommended medical care.
Based on this study, the primary disease was proposed to be the most important indicator of graft survival. In our study, infectious keratitis (bacterial, herpetic, or fungal) was the major cause of second and even in some patients' third graft surgery. In our study, 34 cases had a recurrence of infection. This finding highlights the risk of primary infection recurrence in grafted corneas and emphasizes the importance of antiviral treatments or continuing proper antibiotics after corneal transplantation. Also, a recent cohort study showed that, despite the various changes in the corneal grafting techniques in the past decades, the trephination size and location have no considerable impact on graft survival compared with other factors. However, accurate surgical techniques regarding the location and trephination size help to remove the main focus of infection while keeping a safe margin to the limbal area can reduce the primary infection recurrency in grafted tissue [19].
The quality of donor tissue may also be an influential factor in graft survival. At this tertiary center, the higher quality tissues are preserved for optical keratoplasties [DALK & DSAEK], while moderate quality donor corneas are acceptable for emergency cases and tectonic purposes [e.g., corneal ulcer]. Reasonably, donor tissue with lower quality is prone to fail more easily than higher quality ones. However, this cannot be a convincing reason for reinfection and recurrence.
In our study, diabetes and hypertension were found in 16.2% and 10.8% of the patients; previous studies considered that diabetes is a major risk factor for corneal graft failure [20].
The histopathological evaluation of the regrafting cases showed that the epithelial layer changes were widely observed in accordance with previous studies [20, 21]. Also, following previous studies, the existence of thinning and thickening of the epithelial layer was observed in 32% and 18.9% of regrafts, which may have contributed to corneal graft failure [21]. Another interesting finding in our study was the high number of Sub epithelial bulla in failed graft cases.
Bowman’s layer disorder was the most common histopathological finding in our study (100%), especially, Bowman’s layer Disruption which has been reported in the previous study [21, 22]. Bowman’s layer disorders in our study were associated with graft failure.
Stromal scar and edema in the present study were observed in the majority of cases (81% & 56%), especially in the infectious keratitis associated with corneal graft failure. This histological finding was considered a risk factor for graft failure, and this stromal finding was in accordance with previous studies [20, 21, 22]. Stromal vascularization was observed in 35% of cases, and this finding was observed in previous studies too.
In our study, the inflammatory reaction was graded acute and chronic based on the extent of the inflammatory cell infiltration. The majority of cases had at least some inflammatory reaction within the stroma. Previous studies have shown that newly recruited bone marrow-derived– inflammatory cells produce angiogenic factors and therefore may play role in the mechanisms that induce hemangiogenesis in corneal tissue [23, 24]. Our results about the presence of more stromal vascularization in failed graft tissues that had acute and chronic inflammation may confirm this correlation.
The Descemet layer disruption and retro Descemet fibrous were interesting finding in our study. Also, doubling and excretions of the Descemet layer were observed in some cases and this finding was observed in previous studies [21].
In our study, endothelial cell loss was observed in the majority of cases which is following previous studies [20.21,22]. The endothelial cells are the main target of an immune-mediated attack during a corneal failure reaction [25]. Corneal endothelial cells have no mitotic capacity but have a major role in the process of wound healing by spreading out. Endothelial cells on grafted corneas compared with that of healthy corneas are lost at accelerated rates [25, 26]. Bertelmann et al reported that the mean endothelial cell loss after PK was 28.8% and 39.8%, after 6 and 12 months, respectively [28]. The endothelial cell loss affects the ability of the endothelium to maintain its main function, which eventually leads to a hazy graft [26, 28].
Our study had an important limitation. Despite the high number of regrafting cases, tissue samples were available for histopathological evaluation in limited cases (only 37 cases). This limitation can be due to several reasons, such as a definitive diagnosis of failed graft pathology based on imaging modalities such as confocal and specular microscopy.