Vaccine-Associated Uveitis

Vaccines have had an enormously positive impact on global health, saving an estimated two to three million lives worldwide each year.¹ In 2017 the World Health Organization (WHO) estimated that over 116 million children, representing about nine in ten live births, received at least one vaccination. Even higher infant immunization rates are achieved for many vaccines in North America and Europe. Currently, there are more than two dozen commercially available vaccines, most of which are scheduled specifically for use in infants, children, adolescents, adults or the elderly.² Many more vaccines are in pre-clinical and clinical development.^3,4 At some point in life, therefore, nearly all people on the planet will be vaccinated against common infectious pathogens.

Although uncommon and causality is difficult to prove, there are numerous reports of vaccine-associated adverse events.^5–7 Such events are perhaps most closely monitored by the so-called Vaccine Adverse Event Reporting System (VAERS), a spontaneous post-marketing safety surveillance system jointly established in 1990 by the United States Centers for Disease Control and Prevention (CDC) and Food and Drug Administration (FDA) (https://vaers.hhs.gov/). Data from VAERS together with reports collected by other global surveillance systems is shared with the Uppsala Monitoring Centre (UMC; https://www.who-umc.org/), a WHO Collaborating Centre for International Drug Monitoring. Healthcare professionals, vaccine manufacturers, patients, and patient caregivers can all submit spontaneous reports to these surveillance systems. Such aggregated and publicly available databases play an important pharmacovigilance role, both to provide reassurance of the broad safety of commercially available vaccines and to identify rare, yet possible, vaccine-associated safety signals.⁵ Since inception, more than 500,000 spontaneous reports have been submitted to VAERS.^6,7

Several summaries of vaccine-associated uveitis have appeared over the past decade.^8–15 In the most comprehensive review from 2016,¹⁰ Benage and Fraunfelder identified 289 cases using Medline literature searches and a systematic review of three surveillance systems: the National registry of Drug-Induced Ocular Side Effects (www.eyedrugregistry.com), the UMC, and the VAERS. Vaccination for hepatitis B virus (HBV),¹⁵ either alone or in combination, was reported most frequently (40.5%), followed by immunization for human papillomavirus (HPV; 15.6%),¹² influenza virus (9.7%), Bacille Calmette-Guerin (BCG; 7.3%), measles-mumps-rubella (MMR) alone or in combination (4.8%), varicella virus alone or in combination (4.8%), and hepatitis A virus (HAV) alone or in combination (2.4%). Among the 276 cases where patient gender was reported, 72.1% were female – a proportion similar to the overall VAERS database for those 18 years of age and older.⁷ The mean age at onset was 30 years (range 2 months to 86 years) and the median time from vaccination to onset of uveitis was 16 days (range 1 day to 6 years). Among reports in which treatment and response were provided, uveitis was most often anterior, transient and responded promptly to topical corticosteroids.¹⁰

While vaccines and their adjuvants are designed specifically to interact with and activate the immune system, the precise pathogenesis of vaccine-associated uveitis remains unclear in most patients. Commonly proposed mechanisms have included both molecular mimicry and antigen-specific cell and antibody mediated hypersensitivity reactions. Such mechanisms are not limited to vaccines, however, and have been suggested to play a similarly non-specific role in the pathogenesis of both post-infectious and drug-induced uveitis.^14,16 Of note, two vaccine-specific mechanisms have been suggested and appear to be particularly pertinent. The first mechanism, applicable to live, attenuated vaccines,¹⁷ such as MMR vaccine (M-M-R® II by Merck), nasal spray influenza vaccine (FluMist® by MedImmune/AstraZeneca), rotavirus vaccine (RotaTeq® by Merck and Rotarix® by GlaxoSmithKline), yellow fever vaccine (YF-Vax® by Sanofi) and the original live attenuated varicella vaccine (Zostavax® by Merck), involves the possibility of direct infection by the attenuated, but still active, virus strain. While rare, such infections have been documented, including in the setting of uveitis. Specifically, polymerase chain reaction-based assays have identified persistent replication of the oka/Merck strain following vaccination with that specific virus.¹⁸ The second mechanism involves inflammation induced by one or more adjuvants, typically aluminum salts, routinely used in inactivated or subunit/conjugate vaccines that contain only components or parts of the targeted pathogen.¹⁹ Such autoinflammatory and autoimmune conditions induced by adjuvants (ASIA), known collectively as Shoenfeld syndrome,^19–22 often occur in patients with a family or personal history of autoimmune disease, suggesting a genetic predisposition. Common constitutional symptoms include arthralgia, myalgia, and chronic fatigue.^21,22 Shoenfeld syndrome has been reported most often following immunization against influenza virus, HPV, diphtheria-tetanus-pertussis (DTP), MMR, and BCG. Three letters in this issue of Ocular Immunology & Inflammation (OII) describe uveitis following vaccination.^23–25

Sood et al.²³ described a 43-year-old man of mixed Caucasian and American Indian ancestry who developed bilateral uveitis three days following intradermal HBV. Ocular findings identified 10 days after the onset of symptoms included the presence of both anterior chamber and vitreous inflammation, disc edema, multiple deep yellow-white lesions in the posterior pole and overlying serous retinal detachments (SRD) in each eye. Subsequent multimodal imaging revealed pinpoint leakage and disc edema on fluorescein angiography and hypofluorescent spots in the choroid on indocyanine green angiography. These changes corresponded to the deep yellow-white lesions seen clinically. Optical coherence tomography (OCT) confirmed the presence of SRD and showed both increased choroidal thickness and loss of the normal choroidal vascular architecture – findings suggestive of infiltration by inflammatory cells. Laboratory testing for known causes of uveitis was unrevealing. The patient was treated with combined topical and high-dose oral corticosteroid with complete restoration of vision and resolution of findings. He experienced several recurrences requiring re-treatment, one complicated by the development of peripapillary choroidal neovascularization, and subsequently developed hearing loss, tinnitus, headache and vitiligo leading to the clinical diagnosis of Vogt-Koyanagi-Harada (VKH) syndrome. The authors pointed out that among previously reported cases of HBV-associated uveitis,^10,15 nearly half occurred after the first of what is typically two to four recommended doses, and that several patients experienced a recurrence of their uveitis on re-dosing. While the authors identified no previous reports of VKH-like uveitis following HBV, they did cite a case following influenza vaccine.²⁶ Although not cited, a similar case with VKH-like findings was reported following HPV vaccine.²⁷

Abou-Samra and Tarabishy²⁴ described a 27-year-old woman who developed multiple evanescent white dot syndrome (MEWDS) two weeks following intradermal influenza vaccination. Ocular examination on presentation revealed a central scotoma, mild vitritis and multiple deep retinal lesions involving the posterior pole of the right eye. Fluorescein angiography revealed early hyperfluorescence with late staining of the deep retinal lesions seen clinically. Optical coherence tomography showed that the lesions involved the outer retina. Humphrey visual field testing demonstrated an enlarged blind spot. The patient improved and all findings resolved over eight weeks without treatment. The authors cited three previously published cases of MEWDS following vaccination: one after combined HAV and yellow fever vaccine²⁸; a second following HPV vaccine²⁹; and a third after influenza vaccine.³⁰ Although not cited, MEWDS has also been reported following HBV vaccine by Baglivo et al.,³¹ after HAV vaccine by Fine et al.,³² and following rabies vaccine by Yang et al.³³ Posterior placoid lesions in the spectrum of acute posterior multifocal placoid pigment epitheliopathy (APMPPE) have also been associated with immunization.^34–42 That no fewer than five distinct inactivated or subunit/conjugate vaccines have been associated with MEWDS and APMPPE supports a role for a diverse array of non-infectious antigen triggers in the pathogenesis of these presumed autoinflammatory or autoimmune disorders.

Biancardi and Moraes²⁵ described two women who developed uveitis following fractional intradermal yellow fever vaccine – one a 35-year-old who developed anterior uveitis 10 days after vaccination and the second a 21-year-old woman found to have intermediate uveitis with peripheral vascular leakage on fluorescein angiography two weeks following immunization. Laboratory testing for known causes of uveitis was negative in both patients, and the uveitis resolved in both patients following a six-week taper of topical and oral corticosteroids, respectively. The authors cited previous reports of multifocal choroiditis (MFC) following combined yellow fever, HAV and typhoid vaccine,⁴³ MEWDS after combined yellow fever and HAV vaccine,²⁸ retinal vasculitis after combined yellow fever and Neisseria meningitidis vaccine,⁴⁴ and optic neuritis following combined yellow fever, HAV and HBV vaccine,⁴⁵ but noted that in each of these cases yellow fever was only one of multiple administered vaccines. They cautioned that despite the apparent temporal relationship to the vaccination and the onset of uveitis, together with the fact that vaccination against yellow fever alone was administered in their patient, causality remained presumptive.

Together, these reports highlight the under-appreciated occurrence of intraocular inflammation following intradermal immunization and expand the spectrum of vaccine-associated uveitis. While uveitis has been described most often following the more commonly administered inactivated or subunit/conjugate vaccines - such as those against HBV, HPV, or influenza virus, intraocular inflammation may also follow administration of live, attenuated viruses. Although vaccine-associated uveitis is often anterior, mild, and responsive to topical corticosteroids, more severe forms of posterior and panuveitis, such as MEWDS, APMPPE, and VKH-like syndromes, have also been reported in association with immunization. A recent vaccine history should be elicited, therefore, in all patients who present with uveitis, regardless of anatomical type or severity. For uveitis following immunization with live, attenuated vaccines, providers should consider the possibility of direct infection by the attenuated strain. This can be confirmed by collecting and testing intraocular fluid for the presence DNA specific for the immunized virus. For uveitis following immunization with adjuvant-containing vaccines, patients should be questioned about the occurrence of apparently unrelated autoimmune diseases and about new-onset arthralgias, myalgias, and fatigue – any of which might support a broader, systemic diagnosis of ASIA or Shoenfeld syndrome.

Financial Conflicts

ETC, RSM and FWF have no relevant financial conflicts. MZ is a consultant to AbbVie, Alimera, Santen and Gilead.

Acknowledgments

The authors received no financial support for this editorial.