Incidence and Association of Uveitis with COVID-19 Vaccination: A Systematic Review and Meta-Analysis

ABSTRACT

Purpose

In the wake of the COVID-19 pandemic, vaccines have been pivotal in curbing disease spread and severity. However, concerns over post-vaccination adverse events, including uveitis, an inflammatory ocular condition, have been noted. This systematic review and meta-analysis aimed to evaluate the incidence and association of uveitis following COVID-19 vaccination.

Methods

A literature search was performed across several databases on October 21, 2023. Human studies examining the incidence of uveitis post-COVID-19 vaccination were included. The Newcastle-Ottawa Scale was used for quality appraisal of the included studies. Meta-analysis was performed to assess the overall incidence of uveitis and the relative risk of developing the condition post-vaccination. All statistical analyses were performed using R software version 4.3.

Results

Six studies involving over 2 billion vaccine doses were included. The overall incidence of uveitis was 0.016% (95% CI: 0.010 to 0.026). No significant association was found between vaccination and the onset of uveitis (Relative Risk: 1.45 (95% CI: 0.82 to 2.57, p = 0.12) from four studies. The evidence quality was rated very low due to the limited number of studies and imprecision.

Conclusion

This analysis indicates a low incidence of uveitis following COVID-19 vaccination and no significant association with the vaccine. The findings are constrained by the small number of studies and low certainty of evidence, underscoring the need for further research. Comprehensive and longitudinal studies are necessary to confirm these findings and reinforce public confidence in COVID-19 vaccination programs.

KEYWORDS:

• COVID-19
• meta-analysis
• uveitis
• vaccine

Introduction

The advent of the COVID-19 pandemic, caused by the novel coronavirus SARS-CoV-2, brought about an unparalleled global health crisis.¹ The swift development and deployment of vaccines against COVID-19 have been critical in reducing morbidity and mortality associated with the disease.² However, with mass vaccination campaigns, post-vaccination adverse events have been reported and scrutinized by healthcare professionals, researchers, and the general public.^3,4 One such adverse event that has garnered attention is the onset of uveitis following vaccination. Uveitis is an inflammatory condition affecting the uveal tract of the eye and can lead to significant morbidity, including vision loss if not promptly identified and treated.⁵

The initiation of COVID-19 vaccination programs across the world was a monumental step in the fight against the pandemic. Vaccines, like all pharmacological interventions, carry the potential for side effects, ranging from mild to severe. The majority of vaccine recipients experience minor and transient side effects; however, there is a subset of individuals who develop more serious complications. Uveitis post-vaccination is rare but can be a debilitating condition.⁶ Given the critical function of vision in daily life and the potential for permanent damage, the investigation into the association between COVID-19 vaccines and uveitis is not only a scientific concern but also a public health priority.⁷

The relationship between vaccine administration and the subsequent development of uveitis is complex and not well understood, raising questions about the immune mechanisms at play and the implications for vaccine safety and public health.⁸ The rarity of uveitis as a post-vaccination event requires a robust and systematic approach to capture and analyze these cases, to understand the scope and scale of the issue, and to establish any causal links. Some studies have documented the incidence and clinical trajectory of uveitis in the context of COVID-19 vaccination. For instance, a study underscores the uncommon nature of uveitis post-vaccination, with an incidence of 0.2% and an onset usually occurring within two weeks following vaccination.⁹ This temporal association points towards a possible immunological trigger related to the vaccine. Moreover, the study’s observation of latent infection reactivation in patients with uveitis after vaccination suggests an intricate interplay between the vaccine’s effects and pre-existing latent conditions. The symptoms align with the classical presentation of uveitis and, importantly, respond well to established treatments, resulting in favourable outcomes for the affected patients’ vision.

Even though some systematic reviews are available which summarize the clinical features of uveitis related to COVID-19 based on case series and case reports, no systematic review to date has performed an evaluation of the incidence or association of uveitis due to vaccination quantitatively.^8,10 The present systematic review and meta-analysis aimed to examine the incidence and association of uveitis following COVID-19 vaccination. This is crucial not only for healthcare professionals as they develop guidelines for vaccination and monitor for adverse events but also for maintaining public trust in the vaccines that have become so vital in our efforts to control and end the pandemic.

Methods

This systematic review was performed according to the PRISMA guidelines¹¹ and registered in PROSPERO with registration number: CRD42023477278

Selection criteria

Any type of human studies reporting the incidence or association of cases of uveitis related to COVID-19 vaccination were eligible for inclusion in the review. Studies without the total number of individuals vaccinated, such as individual case reports, were excluded. Animal studies, in vitro studies, reviews, and commentaries were also excluded. All types of vaccines and any doses were considered. Only articles available in English were included. No restrictions were put on the setting, or country where the study was conducted (Table S1).

Literature search

We performed a literature search on October 21, 2023, across various databases, including OVID (Medline), EMBASE, and Web of Science. Keywords and MeSH terms related to COVID-19 and Uveitis were used to build the search strategy. No filters based on study design or language were applied during the search. A detailed search strategy can be seen in Table S2.

Screening

Two independent reviewers were assigned to the screening process, which was conducted in two stages. The primary screening involved reading the title and abstract, followed by full-text screening through thorough reading of the entire text of the article. Any discrepancy regarding the eligibility of studies between the two reviewers was resolved by consultation with a third reviewer. Nested Knowledge software (Nested-Knowledge, MN, USA) was used for deduplication and the screening process.

Data extraction and quality appraisal

For each included study, data were extracted by four reviewers, and cross-checking was performed by another reviewer. The extracted data included the author’s name, year of publication, study design, total number of vaccinated individuals, number of uveitis cases, type of vaccine, and dosage. A quality appraisal for each of the included studies was performed simultaneously by two reviewers. The Newcastle-Ottawa Scale (NOS) was used for quality assessment.¹² The NOS is a tool that provides a systematic approach for evaluating the quality of non-randomized studies, particularly those that assess the outcomes of health interventions. Each study can be awarded a maximum score of 9, with higher scores indicating better quality.¹³

Data synthesis

A meta-analysis was performed to determine the overall incidence of uveitis following COVID-19 vaccination. The total number of uveitis cases and the number of vaccinated individuals were pooled using a random-effects model to estimate the incidence. The incidence rate refers to the proportion of new uveitis cases per 100 vaccine doses administered from 2021 to 2023. Similarly, to find the association between COVID-19 vaccination and uveitis, the pooled Relative Risk (RR) was calculated by considering the number of uveitis cases in the vaccinated and control groups, along with the total sample size in both groups. Following the meta-analysis, a forest plot was generated. To assess the heterogeneity of results across the included studies, the I² statistic was utilized.¹⁴ This statistic represents the percentage of variation across studies that is attributable to heterogeneity rather than chance and ranges from 0% to 100%, with higher values indicating more significant heterogeneity. In addition, the 95% prediction interval was used to understand the variability of outcomes. The tau-squared value, obtained through maximum likelihood estimation, was also calculated to provide a refined measure of between-study variance.^15,16 Statistical significance was determined by a p-value threshold of under 0.05. The statistical package R, version 4.3, was utilized to perform all analyses.¹⁷

Certainty of evidence

The certainty of the evidence was evaluated after synthesizing the results regarding the association of uveitis with COVID-19 vaccination, following the GRADE approach. Overall, the quality of evidence was graded as very low, low, medium, or high.¹⁸

Results

Literature search

A total of 1,621 records were identified through the electronic search from various databases, among which there were duplicates. After removing the duplicates, 1,071 records were screened, and 178 articles were proceeded for full-text screening. A total of 174 records were excluded for various reasons. Finally, 4 studies were found to be eligible. Additionally, 4 articles were identified through citation search, of which 2 were eligible for inclusion. Finally, a total of 6 studies were included in this systematic review. Figure 1 presents the PRISMA flowchart detailing the process of screening and selecting the studies for inclusion.

Figure 1. PRISMA flow diagram showing the screening and selection process.

Characteristics of included studies

The important characteristics of the included studies are summarized in Table 1. Among the studies reviewed, all were retrospective observational studies. The studies were primarily conducted in Israel,¹⁹ Korea,²⁰ Japan,²¹ Singapore,²² and the USA.^7,9 The total cohort size and the number of uveitis cases varied among the studies. Some studies reported a control group along with the vaccinated group, while others did not. The types of vaccines administered were BNT162b2, ChAdOx1, mRNA-1273, Ad26.COV2.S, and BBIBP-CorV. Follow-up period after vaccination were ranged from 14 days to 180 days. The quality of the included studies was generally moderate to high. Two studies^20,22 scored an 7, three^7,9,21 studies scored a 6, and only one¹⁹ study scored a 8 out of 9 on the NOS scale. The detailed summary of quality assessment is presented in Table S3.

Table 1. Characteristics of included studies.

Study	Country	Study design	Follow-up period	Vaccinated group				Control group				No. of Dose	Type of vaccine received
				Age	Males %	Total	No. of uveitis	Total	No. of uveitis cases	Age	Male%
Barda et al.^Citation19	Israel	Retrospective cohort study	42 days	43 (Median)	NA	933,217	26	933,217	20	43 (Median)	NA	Two doses	BNT162b2
Chang et al.^Citation20	Korea	Retrospective cohort study	180 days	54.15	48.37	5,135,028	14,759	2,671,891	3751	33.44	53.05	First dose	ChAdOx1, BNT162b2, mRNA-1273, and Ad26.COV2. S
Chew et al.^Citation9	USA	Retrospective cohort study	14 days	NA	NA	431	6	NA	NA	NA	NA	Two doses	BNT162b2b, BBIBP-CorV vaccine
Hashimoto et al.^Citation22	Singapore	Retrospective cohort study	21 days after the first dose and 84 days after the second dose	NA	NA	Dose 1 = 102846 Dose 2 = 85286	Dose 1 = 19 Dose 2 = 37	Dose 1 = 102846 Dose 2 = 85286	Dose 1 = 12 Dose 2 = 21	70.9 (dose 2); 69.3 (dose1)	NA	Two doses	BNT162b2
Kumar et al.^Citation21	Japan	Retrospective cohort study	NA	43.5	49.2	2305689	254	2305689	266	43.5	49.2	At least one dose	BNT162b2, mRNA-1273, and Johnson Ad26.COV2. S
Singh et al.^Citation7	USA	Retrospective study	NA	46.24	29.43	2061557270	1094	NA	NA	NA	NA	Four doses	BNT162b2, mRNA-1273, and Ad26.COV2. S

Incidence of uveitis in COVID-19 vaccinated individuals

From 6 studies including a total sample size of 2,070,119,767, with 16,195 events the pooled incidence of uveitis with COVID-19 vaccination was found to be 0.016% (95% CI: 0.010 to 0.026), with a heterogeneity of I² = 100%. A prediction interval of 0.0000 to 0.574 was observed. Figure 2 depicts the forest plot of the pooled incidence of uveitis among COVID-19 vaccinations.

Figure 2. Forest plot showing the pooled incidence of uveitis in vaccinated group.

Risk of uveitis following COVID-19 vaccination

From the analysis of four studies, which encompassed a total sample size of 14,660,995 participants, there were 15,095 events reported in the vaccinated group and 4,070 events in the control group. We observed a RR of 1.45 (95% CI: 0.82 to 2.57, p = 0.12), indicating substantial heterogeneity with an I² of 96%. Additionally, a prediction interval ranging from 0.30 to 7.0 was noted. Figure 3 illustrates the forest plot of the meta-analysis. The certainty of the evidence was determined to be very low, as detailed in Table 2.

Figure 3. Forest plot illustrating the relationship between uveitis occurrence and COVID-19 vaccination.

Table 2. Summary of GRADE assessment.

COVID-19 vaccination compared to no vaccination for Causing Uveitis
Certainty assessment							Summary of findings
Participants (studies) Follow-up	Risk of bias	Inconsistency	Indirectness	Imprecision	Publication bias	Overall certainty of evidence	Study event rates (%)		Relative effect (95% CI)	Anticipated absolute effects
							With No vaccination	With COVID-19 vaccination		Risk with No vaccination	Risk difference with COVID-19 vaccination
Uveitis (assessed with: RR)
14660995 (4 observational studies)	not serious	not serious	serious^a	serious^b	none^c	⨁◯◯◯ Very low	4070/6098929 (0.1%)	15095/8562066 (0.2%)	RR 1.145 (0.820 to 2.580)	1 per 1000	0 fewer per 1000 (from 0 fewer to 1 more)

CI: confidence interval; RR: risk ratio.

Explanations.

^aStudies were not conducted to directly assess Association

^bConfidence Intervals overlap with null effect

^cDoi plot showed asymmetry with high LFK index values

Sensitivity analysis and publication bias

We performed leave-one-out analysis by removing one study at a time and reanalysis for the association of uveitis with COVID-19 vaccination. When the study by Kumar et al. 2023 was omitted, the RR increased to 1.89 (95% CI: 0.82 to 2.57) (Figure 4). Due to the limited number of studies, publication bias was not assessed.

Figure 4. Sensitivity analysis of relationship between uveitis occurrence and COVID-19 vaccination.

Discussion

This is the first systematic review and meta-analysis to synthesize results on the incidence and relationship between occurrence of uveitis following COVID-19 vaccination. There is a spectrum of adverse events related to the COVID-19 vaccine reported in the literature; uveitis is one of the many adverse events. Our analysis showed that the incidence of uveitis is very rare following COVID-19 vaccination. Even though we found statistical heterogeneity among studies for the assessment of incidence in the meta-analysis, all the included studies revealed a very low incidence (<0.1%). The heterogeneity could be due to the nature of the studies or variation in the rate of reporting of adverse effects. We could not find any significant association between the incidence of uveitis following COVID-19 vaccination in our analysis. The certainty of the evidence was also found to be very low as per the GRADE criteria. Although there are several case reports and case series that present the clinical manifestations of uveitis related to COVID-19 and vaccination, longitudinal and quantifiable cross-sectional studies are very limited in the current existing literature.

Our study is the first to quantify occurrences of vaccine-associated uveitis (VAU) following COVID-19 vaccination, expanding on previous reviews that outlined its clinical presentations. For instance Cherif et al. identified ocular redness and visual impairment as predominant symptoms, with fewer cases reporting floaters and vision loss.⁸ The acute presentation was more common, typically affecting the eye’s anterior segment. Interestingly, cases of panuveitis were twice as prevalent as those of posterior uveitis. A significant number of patients reported experiencing uveitis for the first time after vaccination, indicating a potential link. The condition was primarily inflammatory and responded well to treatment, with complete resolution in most instances. Complications were rare, with transient increases in intraocular pressure (IOP) being the most frequent.

The study by Singh and colleagues revealed that around three-quarters of the individuals in their study developed VAU within one month of receiving the COVID-19 vaccine.⁷ Notably, a higher number of VAU cases occurred after the initial vaccine dose (41.32%) than after the second dose (34.1%). Tomkins-Netzer et al. reported that over half of their subjects (52%) had preexisting uveitis and that a significant majority (90.96%) experienced anterior uveitis post-vaccination. According to data from the Vaccine Adverse Event Reporting System (VAERS), a small fraction of VAU cases were in individuals with a prior uveitis diagnosis (9.7%) or with systemic autoimmune conditions (1.2%). Furthermore, less than half (44.9%) of these reported cases were diagnosed with anterior uveitis subsequent to vaccination. Another study conducted by Chang et al. noted a rise in non-infectious uveitis cases up to six months following vaccination.²⁰ Their findings indicate a relatively higher association of non-anterior uveitis with vaccination compared to anterior or total uveitis. In particular, an elevated risk of developing non-anterior uveitis was identified among female subjects, with a history of non-infectious uveitis being the predominant risk factor. In their nationwide cohort study, when comparing vaccinated individuals to a control group of unvaccinated and uninfected individuals, there was a slight uptick in non-anterior uveitis risk post-vaccination, predominantly in women. However, despite utilizing multivariate Cox regression to adjust for potential confounders, residual bias might persist due to marked differences between the vaccinated and non-vaccinated groups, along with the rarity of non-anterior uveitis in those who were unvaccinated. In contrast, a retrospective population-based study did not find an elevated risk of non-infectious uveitis (NIU) following COVID-19 vaccination among patients with no previous history of uveitis, a conclusion supported by the Self-Controlled Case Series (SCCS) analysis.²¹ This study’s findings are in line with previously reported NIU incidence rates in the USA, which are estimated to be between 24.9 and 52.4 per 100,000 person-years, thereby reinforcing the credibility of the study’s methodology. Additionally, the analysis of the matched cohort showed an observable increase in NIU risk post-vaccination, which appeared to be linked with older age and the use of immunomodulatory therapy. The authors mention that these factors are generally acknowledged as risk factors for NIU due to the immune system irregularities often present in older individuals or those with an underlying immune-mediated condition necessitating immunosuppressive treatment.

The studies included in our analysis did not clearly differentiate between the types of uveitis outcomes following vaccination except the study by Chew et al.²³ Specifically, it was ambiguous whether the uveitis cases were new instances triggered by vaccination or flare-ups of pre-existing conditions. While some studies acknowledged a history of uveitis in a subset of patients, they did not provide detailed information on the nature of these outcomes, specifically whether they were de novo occurrences or exacerbations of existing uveitis. Notably, the overall incidence of uveitis observed in our study was relatively low. This suggests that, despite the limitations in distinguishing between new and existing cases, the history of uveitis has a minimal impact on the general conclusions of our analysis.

The underlying mechanisms of VAU are not fully understood; however, it is hypothesized that autoimmune responses triggered by vaccination play a central role.²⁴ This autoimmune activity could be a result of various processes, including molecular mimicry, where vaccine-induced antibodies cross-react with ocular tissue, and hypersensitivity reactions, possibly exacerbated by vaccine adjuvants.^25,26 Vaccinations are known to initiate an inflammatory response through the stimulation of type 1 interferon expression, which helps mount the body’s immune defense. Yet, this process can sometimes lead to the inadvertent production of autoantibodies, inciting an autoimmune response.^25,27 Rabinovitch and colleagues proposed that mRNA vaccines might cause a type I auto immune reaction, characterized by an increased production of type 1 interferon.²⁸ Similarly, Cunningham and associates suggest a type IV hypersensitivity reaction, potentially caused by molecular mimicry between peptides found in the uveal tissue and those present in vaccines.²⁹ These hypotheses, while providing a foundation for understanding VAU post-COVID-19 vaccination, require further research for confirmation.

Despite the low incidence of VAU following COVID-19 vaccination, clinicians should remain vigilant for ocular symptoms in vaccinated individuals. Awareness of this possible side effect can lead to prompt diagnosis and management, potentially mitigating the risk of severe outcomes. It is crucial that healthcare providers are aware of the symptoms associated with VAU such as redness, visual impairment, floaters, and vision loss – to ensure early intervention and referral to ophthalmologists when necessary. Additionally, the higher incidence of VAU in females and those with a history of non-infectious uveitis reported in some studies suggests that patient counseling and follow-up may be warranted. This would be especially relevant for individuals with a preexisting history of uveitis or systemic autoimmune conditions, who appear to be at a heightened risk. Physicians should consider discussing the risks and benefits of COVID-19 vaccination with these patients, taking into account their personal health history and potential risk factors for VAU. Healthcare systems might also need to consider establishing protocols for the surveillance of post-vaccination adverse events, including VAU. This could involve regular check-ups or establishing reporting mechanisms that could alert healthcare providers to possible cases of VAU. Incorporating questions about recent vaccinations in the diagnostic work-up for patients presenting with uveitis could help in identifying vaccine-related cases. Policymakers and public health officials could use the results of this review to inform vaccine policy. While the benefits of COVID-19 vaccination in controlling the pandemic are well-established, understanding and communicating potential risks are essential for maintaining public trust. Transparency regarding the rarity of VAU and other adverse events, while emphasizing the overall safety and efficacy of vaccines, is important for informed consent and vaccination uptake. In the context of vaccine hesitancy, it is crucial to balance the dissemination of information about possible adverse effects with the clear communication of the comparative risk of complications from the COVID-19 disease itself. The data could also be used to guide vaccine development. Understanding the pathophysiological mechanisms behind VAU could lead to the refinement of vaccine components to minimize autoimmune reactions. Collaborations between vaccine manufacturers, researchers, and clinicians can enhance the safety profile of vaccines.

This systematic review and meta-analysis represent the first concerted effort to quantitatively assess the overall incidence and relationship between occurrence of uveitis following COVID-19 vaccination. We identified a significant research gap, specifically the lack of longitudinal studies in this area. Studies didn’t make a distinction between the reported uveitis, whether it is de novo or flare of previous history of uveitis. Our review was limited to articles published in English, which is a constraint that may have excluded relevant studies published in other languages. Additionally, the studies available for inclusion were limited to certain countries, which may not provide a global perspective on the incidence of vaccine-associated uveitis. The number of studies that met our inclusion criteria was relatively small, and the certainty of the evidence was determined to be very low. There is a clear need for further research to address these gaps.

Conclusion

The incidence of uveitis associated with COVID-19 vaccination was found to be low in the current literature. Despite the limited number of studies and the low certainty of evidence, the findings suggest no significant association between COVID-19 vaccines and the onset of uveitis. Further comprehensive and longitudinal research is essential to enhance our understanding of vaccine-related adverse events and to strengthen trust in the efficacy of vaccination programs.

Authors contributions

• Substantial contribution to the conception or design of the work: A.S.A, H.A, N.A.A.K, M.F.A.S, D.M, A.D

• Acquisition of data: T.S, A.A.R, M.N.K, SG, J.J.B

• Analysis and interpretation of data for the work: H.A.A, T.S, A.A.R, M.N.K, SG, Q.S.Z

• Drafting the work: J.K.G, K.B.P, P.S, D.M, A.D

• Revising it critically for important intellectual content: A.S.A, H.A, N.A.A.K, M.F.A.S, H.A.A, T.S, J.J.B

• Resources: M.N.K, SG, Q.S.Z, J.K.G

• Final approval of the version to be published: All authors (A.S.A, H.A, N.A.A.K, M.F.A.S, H.A.A, T.S, A.A.R, M.N.K, SG, Q.S.Z, J.K.G, D.M, A.D, J.J.B, P.S)

Acknowledgments

The authors would like to thank the developers of Nested Knowledge software for providing the software.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/09286586.2024.2343714.

Additional information

Funding

This work was supported by Qatar National Library funded the publication of this article.