Characteristics and Clinical Ocular Manifestations in Patients with Acute Corneal Graft Rejection after Receiving the COVID-19 Vaccine: A Systematic Review

This study aimed to determine the characteristics and clinical ocular manifestations of acute corneal graft rejection after coronavirus disease 2019 (COVID-19) vaccination. We conducted an online search of the PubMed and EMBASE databases. Data on recipients’ characteristics, corneal transplantation types, interval between vaccination and allograft rejection, clinical manifestations, and graft rejection medication were extracted. Thirteen articles on 21 patients (23 eyes) with acute corneal graft rejection after COVID-19 vaccination, published between April and December 2021, were included. The median (interquartile range) age at the onset of rejection was 68 (27–83) years. Types of transplantation included penetrating keratoplasty (12 eyes), Descemet membrane endothelial keratoplasty (six eyes), Descemet stripping automated endothelial keratoplasty (four eyes), and living-related conjunctival-limbal allograft (one eye). The interval between vaccination and rejection ranged from 1 day to 6 weeks. Corneal edema was the leading clinical manifestation (20 eyes), followed by keratic precipitates (14 eyes) and conjunctival or ciliary injection (14 eyes). Medications included frequently applied topical corticosteroids (12 eyes), followed by a combination of topical and oral corticosteroids (four eyes). In addition, the clinical characteristics of corneal allograft rejection after COVID-19 vaccination were identified. Corneal transplant recipients may require further vaccination, necessitating appropriate management and treatment.


Introduction
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) caused an unprecedented worldwide pandemic of the coronavirus disease 2019 (COVID-19) [1,2]. On 10 March The primary aim of this study was to systematically evaluate and characterize the currently reported cases of acute corneal graft rejection after COVID-19 vaccine administration. In particular, we focused on variables such as age, sex, and ethnicity of patients; type of vaccine; the number of vaccinations; type of corneal transplantation; the interval between corneal transplantation and rejection; ocular findings; the interval between vaccination and rejection; and medications used to treat rejection.
Search results were compiled using Endnote software X9.3.3 (Clarivate Analytics, Philadelphia, PA, USA). Two independent researchers (K.F. and T.I.) screened the retrieved articles in accordance with the defined quality standards for reporting systematic reviews and meta-analyses for observational studies [25]. Additionally, the same investigators independently assessed the full texts of eligible articles to reach a consensus.

Risk of Bias Assessment
The risk of bias in the individual studies was assessed using either the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Case Reports or the JBI Critical Appraisal Checklist for Case Series [26]. The checklists for case reports and case series consist of 8 and 10 items, respectively, with response choices of "yes", "no", "unclear", or "not applicable". Two investigators (K.F. and K.N.) independently assigned an overall risk of bias to each eligible study, and if they disagreed, a third reviewer (T.I.) was consulted. The risk of bias was determined considering the total number of "yes" responses, with ≥70%, 50-69%, and ≥49% of the responses indicating low, moderate, and high risk of bias, respectively [27].

Data Extraction
Two independent reviewers (K.F. and T.I.) extracted data from each eligible article using a standardized data extraction sheet and then cross-checked the results. Disagreements between the reviewers regarding extracted data were resolved through discussion with a third reviewer (J.S.). The following information was extracted: first author's name; date of publication; type of study (case report and case series); country; characteristics of patients with acute corneal graft rejection after COVID-19 vaccination, including age, sex, and race; type of vaccine; the number of vaccinations; type of corneal transplantation; the interval between corneal transplantation and rejection, the interval between vaccination and rejection; ocular findings; and medications.

Statistical Analysis
Data analyses were performed considering the Updated Method Guidelines for Systematic Reviews in the Cochrane Collaboration Back Review Group [28].

Discussion
Since the beginning of the COVID-19 pandemic in 2019, efforts toward vaccination have continued worldwide due to the uncontrollable spread of SARS-CoV-2 infection and the increasing number of COVID-19 cases. However, despite the low occurrence rates of high-risk complications after vaccination, an appreciable number of vaccine recipients have experienced a wide range of post-vaccination symptoms. At present, COVID-19 vaccines are believed to be associated with a spectrum of systemic symptoms, and appropriate interventions should be undertaken on a case-by-case basis. In this systematic review, we extracted data regarding clinical features of acute corneal allograft rejection (21 patients [23 eyes]) that occurred after the administration of the COVID-19 vaccine. Among them, >95% of eyes (22/23 eyes) had corneal allograft rejection within 3 weeks from vaccination (mean: 10.4 days, median: 7 days). As the global society promotes additional booster schedules in consideration of emerging variants, it is empirical that the effects of these vaccines on corneal grafts be elucidated. To establish appropriate immune-modulatory interventions, continued data accrual and investigation of the effects of COVID-19 vaccines should be conducted by observing the vaccination course and associated physiological changes in corneal allograft recipients.
In our analysis of 21 patients (23 eyes) with signs of corneal allograft rejection after receiving COVID-19 vaccines, the median interval between corneal transplantation and graft rejection after COVID-19 vaccination was 2 years. Two of these patients (two eyes) underwent PKP > 20 years ago and had no history of acute or chronic corneal allograft rejection [33,38]. The long-term stability and temporariness of vaccination and allograft rejection suggest that immune responses to the vaccine may have played a role in transplant rejection. Two separate studies investigated the possibility of confounding viral infections, including herpes simplex and varicella-zoster viral infections, which are known risk factors for corneal allograft rejection after PKP or DMEK [20,38], through anterior-chamber, aqueous-humor polymerase chain reaction (PCR). However, PCR results for confounding viral infections were negative in both patients. Additionally, two cases (four eyes) of post-DMEK and -DSAEK bilateral corneal allograft rejection after vaccination were reported. These cases suggest a systemic inflammatory etiology for corneal graft rejection [33,38]. The detailed pathophysiology of the relationship between COVID-19 vaccination and corneal allograft rejection remains unclear. However, the present study revealed cases with minimal confounding variables and appreciable temporal correlation with COVID-19 vaccination, positing the vaccine's role in inducing acute corneal allograft rejection.
Another risk factor for vaccine-associated corneal allograft rejection may be the presence of a corneal bed with a high rejection risk. Recurrent infections, autoimmune disease complications, and multiple corneal transplantations lead to progressive neovascularization and lymphangiogenesis of the cornea, virtually eliminating the immune privilege of the anterior eye [64]. Such corneal beds have a 40-90% rejection rate in subsequent transplantations. Of the 23 eyes included in this review, nine had undergone more than one transplantation [29,31,[35][36][37]39,40]. Rallis et al. reported a case of an acute corneal allograft rejection following BNT162b2 vaccine administration with a surgical history of DSAEK and a re-do PKP for an existing Fuchs' corneal endothelial dystrophy [31]. This suggests that the angiogenesis and lymphangiogenesis induced by repeated insults to the cornea (e.g., repeated corneal allograft transplantation) may predispose the corneal bed to the high immune stress that follows COVID-19 vaccine administration. Consequently, high-risk allograft recipients who have undergone repeated corneal transplants should be monitored and thoroughly examined after vaccination. Nine of the eyes included in this review developed corneal graft failure after vaccination [29,31,[35][36][37]39,40], all of which had undergone repeated corneal transplants. As inadequate control of the corneal immune activity may subject the allograft to inflammatory insult, even from minor vaccinations, owing to its newly developed systemic communication, high-risk allograft recipients should be frequently followed up with appropriate immune-suppressive management.
Interestingly, acute cases of graft rejection of organs other than the cornea after the administration of the COVID-19 vaccines have been relatively rare, with only one case reported to date [65]. Excluding one patient (one eye) from the total reviewed patients [34], none of the patients received oral steroids or immune-modulatory medications. This contrasts with other organ transplant cases, where lifelong immunosuppressive or steroid therapy is typically prescribed [66]. Furthermore, the dampened inflammatory responses to the vaccine may have reduced the immune stress on these grafted organs. Therefore, prescribing oral immunomodulators or increasing the frequency of topical steroid administration should be considered for allograft recipients with a high rejection risk because of continued angiogenesis or lymphangiogenesis, particularly from 7 to 28 days after COVID-19 vaccine administration, when immune responses are at their peak [58].
This systematic review has certain limitations. First, the number of reported cases reviewed in this study is limited because of the recency of the ongoing pandemic. COVID-19 vaccines were approved and made clinically available in December 2020; therefore, only 21 cases with corneal allograft rejection after receiving the COVID-19 vaccine have been reported. Consequently, this study did not conduct a meta-analysis for the specific outcomes. Second, the generalizability of the results should be considered with caution because the studies included are from eight different countries with an ethnically white-predominant subject pool. There are no reports on patients from East Asia, and future investigations should aim to accrue cases from various geographical and ethnic backgrounds. Finally, the results of this systematic review do not elucidate the detailed pathophysiology of acute allograft rejection. Although two of the included studies performed anterior chamber aqueous humor PCR [20,38], the remaining studies did not successfully rule out any confounding causes of corneal graft rejection. In addition, this study included patients who experienced corneal graft failure >3 weeks after vaccination [40]. The COVID-19 vaccination may not have had direct links to these reported cases of corneal graft rejection. Therefore, the 21 cases of graft failure included in this review may have confounding aspects beyond the direct effects of the COVID-19 vaccine, including inadequate immune suppression at the time of vaccination. Future studies should incorporate extensive examinations-such as corneal endothelial cell density and viral panels-to remove the effects of confounders that may cause corneal graft failure and isolate vaccine-induced corneal changes. Nonetheless, the cases of two patients with a stable post-transplantation course over 20 years who developed signs of rejection after receiving the COVID-19 vaccine [33,38] and the bilaterality of some included cases suggest that the systemic inflammation and immune system upregulation caused by the COVID-19 vaccine may be associated with acute corneal allograft rejection [33,38].
In conclusion, this systematic review identified clinical features and host factors associated with corneal allograft rejection after COVID-19 vaccination. As the virus continues to spread, additional booster COVID-19 vaccine schedules are expected. Therefore, proper follow-up of corneal allograft recipients and interventions to prevent corneal allograft rejection after administering the COVID-19 vaccine may be crucial.